Discussion:
[avr-libc-dev] [bug #50270] Article "Problems with reordering code" misleading
anonymous
2017-02-09 22:07:36 UTC
Permalink
URL:
<http://savannah.nongnu.org/bugs/?50270>

Summary: Article "Problems with reordering code" misleading
Project: AVR C Runtime Library
Submitted by: None
Submitted on: Thu 09 Feb 2017 10:07:34 PM UTC
Category: Documentation
Severity: 3 - Normal
Priority: 5 - Normal
Item Group: Documentation
Status: None
Percent Complete: 0%
Assigned to: None
Originator Email: ***@onet.pl
Open/Closed: Open
Discussion Lock: Any
Release: 2.0.0
Fixed Release: None

_______________________________________________________

Details:

The article available here:
http://www.nongnu.org/avr-libc/user-manual/optimization.html#optim_code_reorder/optimization_1optim_code_reorder.html
contains an essential mistake and is misleading for avr-gcc users.

In short: the author of the article in online documentation makes some
analysis of a tricky case of using memory barrier in gcc and makes false
conclusions. Quote:

"To sum it up:

memory barriers ensure proper ordering of volatile accesses

memory barriers don't ensure statements with no volatile accesses to be
reordered across the barrier"

Point 1 makes no sense - volatile accesses are ensured to be properly ordered
without any barriers - this is guaranteed by C standard.

Point 2 is false - a memory barrier ensures proper ordering of all, including
non-volatile, memory accesses. This covers all global data, but not
necesserily local data which can be placed in registers. In the analyzed case,
variable val is a local variable. This is the real reason why memory barrier
does not ensure strict ordering of operations on val. Memory barriers won't
influence management of local variables stored in registers And this is
perfectly fine, as only variables stored in memory can be accessed from
various CPU contexts.

So the sentence "Unfortunately, at the moment, in avr-gcc (nor in the C
standard), there is no mechanism to enforce complete match of written and
executed code ordering" is correct. But the further conclusions (the two
points quoted earlier) are incorrect and strongly misleading.




_______________________________________________________

Reply to this item at:

<http://savannah.nongnu.org/bugs/?50270>

_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/
Joerg Wunsch
2017-02-10 09:03:53 UTC
Permalink
Follow-up Comment #1, bug #50270 (project avr-libc):

Your criticism seems to confuse volatile memory access with
volatile asm statements. Given the matter is known to be
tricky, it would have been better to subscribe to the
avr-libc mailing list, and discuss the wording there with
the other developers (including the author of that snippet),
rather than immediately declaring it a "bug".

After all, that article has been written for a reason, after
certain observations have been analyzed and discussed prior
in Internet forums and mailing lists.

As it is now, even after re-reading it, the wording and
examples of the article still look much more reasonable to
me than your blunt statement "it cannot be what is not
supposed to be".

_______________________________________________________

Reply to this item at:

<http://savannah.nongnu.org/bugs/?50270>

_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/
Marcin Godlewski
2017-02-10 11:06:23 UTC
Permalink
Follow-up Comment #2, bug #50270 (project avr-libc):

1. I don't think I confuse volatile memory access with volatile asm
statements.What is your point here?

2. I've already discussed the problem in the list avr-gcc-***@nongnu.org and
I got feedback to submit a bug report. Please search for a topic
Avr-libc-user-manual: "Problems with reordering code".

3. I don't say the article is written without a reason. But it simply seems to
give inaccurate conclusions.

4. I also don't understand your last point. What blunt statement did I give? I
simply explained why I think the conclusions in article are wrong. I
understand that something may "look much more reasonable" to you, but such
comments doesn't bring any value in the discussion. I hope to have a
content-related discussion.


_______________________________________________________

Reply to this item at:

<http://savannah.nongnu.org/bugs/?50270>

_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/
Joerg Wunsch
2017-02-10 14:06:14 UTC
Permalink
Follow-up Comment #3, bug #50270 (project avr-libc):

The bug report is fine for a clear bug. However, for
something to be discussed first, it's a poor medium: if
the author of that text (who is on the developers
mailinglist) wants to reply something, he stands no chance
to add it to the bug report. If we discuss it on the list,
you'll miss the replies.

If you've got a suggestion for a better explanation and/or
wording of the phenomenon shown there, please tell us so.
The entire purpose of the text is to make users aware that
there are potential pitfalls in optimization which at least
currently cannot be controlled by any compiler statement.

_______________________________________________________

Reply to this item at:

<http://savannah.nongnu.org/bugs/?50270>

_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/
Marcin Godlewski
2017-02-10 14:32:18 UTC
Permalink
Follow-up Comment #4, bug #50270 (project avr-libc):

I'm sorry if I have chosen the wrong medium. I tried to point out the problem
on a mailing list already, but hardly anyone was interested to discuss it. I
may have chosen the wrong mailing list, but this was the only one I found.

But I still don't get you. If the author or anybody else wants to discuss the
problems reported by me, they are free to do it here - in the discussion of a
bug report.

I have provided an explanation of the problem that was - I believe - wrongly
explained in the article. Surely my explanation is not perfectly worded.

And to be clear, my intention is to improve online documentation of avr-libc
to make it more useful for everybody. If you have any remarks regarding the
subject, please provide them. Discussing whether this should be a bug report
or a mail on this or that mailing list, without any points regarding the
subject of the report, is really not the right direction.


_______________________________________________________

Reply to this item at:

<http://savannah.nongnu.org/bugs/?50270>

_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/
Marcin Godlewski
2017-02-10 14:36:57 UTC
Permalink
Follow-up Comment #5, bug #50270 (project avr-libc):

As I can see, all the discussion is automatically sent to
avr-libc-***@nongnu.org, so I guess the author of the text can see the
discussion and join it. Isn't this right?

_______________________________________________________

Reply to this item at:

<http://savannah.nongnu.org/bugs/?50270>

_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/
Joerg Wunsch
2017-02-10 14:46:46 UTC
Permalink
Follow-up Comment #6, bug #50270 (project avr-libc):

Yes, he can discuss it there, but he cannot append anything
to the bug report itself.

_______________________________________________________

Reply to this item at:

<http://savannah.nongnu.org/bugs/?50270>

_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/

Loading...