unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
* disable automatic native-compilation?
@ 2022-07-09 16:49 Stephen Leake
  2022-07-09 17:23 ` Eli Zaretskii
                   ` (2 more replies)
  0 siblings, 3 replies; 18+ messages in thread
From: Stephen Leake @ 2022-07-09 16:49 UTC (permalink / raw)
  To: emacs-devel

I've just started using Emacs 28. I'm trying run the ada-mode test
suite, which lauches a new Emacs instance for each test.

Now that instance also starts a native-compile process, which slows
things down (a lot), and does not terminate cleanly when the Emacs
instance does at the end of the test. I'm on Windows, so this pops up a
dialog that I have to acknowledge eventually.

The elisp manual section on native compilation does not say how to
disable it. I managed to disable starting the native-compile process by:

(defun native-comp-available-p () nil)

That sometimes causes an error about "can't find libgccjit", which
aborts the test. I can't track down the specific cause of that error,
because even C-h f foo gives the "can't find libgccjit" error once it
happens.

I tried M-x apropos-variable native, which turned up some interesting
possibilities:

native-comp-deferred-compilation

The documentation for this variable says:
    "If non-nil compile loaded .elc files asynchronously."

what happens if this is nil? the files are compiled synchronously, or
not at all?

Setting that nil instead of redefining native-comp-available-p seems to
work.


So, I suggest:

- There should be a simple way to completely disable native compilation;
Emacs should never do anything that I can't control. If setting
native-comp-deferred-compilation nil is the right way to do this, it
should be documented in the native compilation section of the elisp
manual.

- When native-comp-available-p returns nil, there should be no errors
about libgccjit

-- 
-- Stephe



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-09 16:49 disable automatic native-compilation? Stephen Leake
@ 2022-07-09 17:23 ` Eli Zaretskii
  2022-07-10 15:10   ` Lynn Winebarger
  2022-07-09 22:36 ` Tor Kringeland
  2022-08-06 15:32 ` Lynn Winebarger
  2 siblings, 1 reply; 18+ messages in thread
From: Eli Zaretskii @ 2022-07-09 17:23 UTC (permalink / raw)
  To: Stephen Leake; +Cc: emacs-devel

> From: Stephen Leake <stephen_leake@stephe-leake.org>
> Date: Sat, 09 Jul 2022 09:49:59 -0700
> 
> I've just started using Emacs 28. I'm trying run the ada-mode test
> suite, which lauches a new Emacs instance for each test.
> 
> Now that instance also starts a native-compile process, which slows
> things down (a lot), and does not terminate cleanly when the Emacs
> instance does at the end of the test.

How does it not "terminate cleanly"? what uncleanliness is left?
Please do tell the details.

> I'm on Windows, so this pops up a dialog that I have to acknowledge
> eventually.

What dialog did pop up that you needed to acknowledge? what did it
say?

> The elisp manual section on native compilation does not say how to
> disable it.

That's because we don't see a need for disabling that, in a build that
supports native compilation.  The thought was that people who don't
want native compilation will build Emacs without it in the first
place.

> native-comp-deferred-compilation
> 
> The documentation for this variable says:
>     "If non-nil compile loaded .elc files asynchronously."
> 
> what happens if this is nil? the files are compiled synchronously, or
> not at all?

You will need to compile them by hand in that case.  Automatic native
compilation always works asynchronously, because otherwise it would
have wedged your session for many seconds, something that is
unthinkable.

> So, I suggest:
> 
> - There should be a simple way to completely disable native compilation;
> Emacs should never do anything that I can't control.

I don't necessarily agree.  Emacs doesn't do here anything different
from what it did before: it loads a compiled Lisp library when you
request that.  It only compiles it to native code the first time it is
loaded when no .eln file exists for it, that's all.  But that's a
side effect of loading the natively-compiled Lisp that you requested
in a build that _you_ built with natively-compilation support.

It sounds like you encountered problems in some very special
situation, where you are running tests using sub-processes, and your
test suite was not ready for native-compilation.  So I'm not sure we
should draw far-fetching conclusion from your use case.  But whether
this impression of mine is correct can only be established by
discussing the details of your scenario, which you didn't provide.  I
suggest to submit a detailed bug report.

> If setting native-comp-deferred-compilation nil is the right way to
> do this, it should be documented in the native compilation section
> of the elisp manual.

That variable is not a user variable, it exists for internal use
purposes.  The ELisp manual does document how to disable
native-compilation of a specific file (similarly to how you disable
byte-compilation of a file).

> - When native-comp-available-p returns nil, there should be no errors
> about libgccjit

Please tell the details and present a reproducible recipe (preferably
in a bug report), then we can discuss what happens in that specific
situation.  In general, you are not supposed to override that function
unless you know what you are doing.



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-09 16:49 disable automatic native-compilation? Stephen Leake
  2022-07-09 17:23 ` Eli Zaretskii
@ 2022-07-09 22:36 ` Tor Kringeland
  2022-07-10  8:58   ` Philip Kaludercic
  2022-07-11 12:54   ` Andrea Corallo
  2022-08-06 15:32 ` Lynn Winebarger
  2 siblings, 2 replies; 18+ messages in thread
From: Tor Kringeland @ 2022-07-09 22:36 UTC (permalink / raw)
  To: Stephen Leake, emacs-devel

Stephen Leake <stephen_leake@stephe-leake.org> writes:

> - There should be a simple way to completely disable native compilation;
> Emacs should never do anything that I can't control. If setting
> native-comp-deferred-compilation nil is the right way to do this, it
> should be documented in the native compilation section of the elisp
> manual.

Doesn't setting

  (setq native-comp-speed -1)

disable native compilation?


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-09 22:36 ` Tor Kringeland
@ 2022-07-10  8:58   ` Philip Kaludercic
  2022-07-11 12:54   ` Andrea Corallo
  1 sibling, 0 replies; 18+ messages in thread
From: Philip Kaludercic @ 2022-07-10  8:58 UTC (permalink / raw)
  To: Tor Kringeland; +Cc: Stephen Leake, emacs-devel

Tor Kringeland <tor.kringeland@ntnu.no> writes:

> Stephen Leake <stephen_leake@stephe-leake.org> writes:
>
>> - There should be a simple way to completely disable native compilation;
>> Emacs should never do anything that I can't control. If setting
>> native-comp-deferred-compilation nil is the right way to do this, it
>> should be documented in the native compilation section of the elisp
>> manual.
>
> Doesn't setting
>
>   (setq native-comp-speed -1)
>
> disable native compilation?

The docstring says it does, but it certainly is not an intuitive option
if you are looking for a way to disable native comp.



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-09 17:23 ` Eli Zaretskii
@ 2022-07-10 15:10   ` Lynn Winebarger
  2022-07-10 15:39     ` Eli Zaretskii
  0 siblings, 1 reply; 18+ messages in thread
From: Lynn Winebarger @ 2022-07-10 15:10 UTC (permalink / raw)
  To: Eli Zaretskii; +Cc: Stephen Leake, emacs-devel

On Sat, Jul 9, 2022 at 1:26 PM Eli Zaretskii <eliz@gnu.org> wrote:
>
> > From: Stephen Leake <stephen_leake@stephe-leake.org>
> > Date: Sat, 09 Jul 2022 09:49:59 -0700
> >
> > I've just started using Emacs 28. I'm trying run the ada-mode test
> > suite, which lauches a new Emacs instance for each test.
> >
> > Now that instance also starts a native-compile process, which slows
> > things down (a lot), and does not terminate cleanly when the Emacs
> > instance does at the end of the test.
>
> How does it not "terminate cleanly"? what uncleanliness is left?
> Please do tell the details.
>
> > I'm on Windows, so this pops up a dialog that I have to acknowledge
> > eventually.
>
> What dialog did pop up that you needed to acknowledge? what did it
> say?
>
If Stephen's building native compilation under cygwin, perhaps he's
referring to an error message about not being able to load the ELN
file, either in the form of a "fork" failure, or an explicit complaint
about the dll base address conflicting with another dll that's already
been loaded.  For a 64-bit build, a restart of emacs seems to be
enough to cure the issue, although I have also had issues that didn't
seem to be resolved without doing a "rebase all".
I don't know if this behavior is considered a bug (in either libgccjit
or emacs) or just a basic limitation of using the cygwin platform.
It is definitely an annoyance when either installing packages or when
one of the package's features/modules is autoloaded the first time.

Lynn



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-10 15:10   ` Lynn Winebarger
@ 2022-07-10 15:39     ` Eli Zaretskii
  2022-07-10 18:32       ` Lynn Winebarger
  0 siblings, 1 reply; 18+ messages in thread
From: Eli Zaretskii @ 2022-07-10 15:39 UTC (permalink / raw)
  To: Lynn Winebarger, Ken Brown; +Cc: stephen_leake, emacs-devel

> From: Lynn Winebarger <owinebar@gmail.com>
> Date: Sun, 10 Jul 2022 11:10:27 -0400
> Cc: Stephen Leake <stephen_leake@stephe-leake.org>, emacs-devel <emacs-devel@gnu.org>
> 
> If Stephen's building native compilation under cygwin, perhaps he's
> referring to an error message about not being able to load the ELN
> file, either in the form of a "fork" failure, or an explicit complaint
> about the dll base address conflicting with another dll that's already
> been loaded.  For a 64-bit build, a restart of emacs seems to be
> enough to cure the issue, although I have also had issues that didn't
> seem to be resolved without doing a "rebase all".
> I don't know if this behavior is considered a bug (in either libgccjit
> or emacs) or just a basic limitation of using the cygwin platform.
> It is definitely an annoyance when either installing packages or when
> one of the package's features/modules is autoloaded the first time.

This was discussed with Ken Brown (CC'ed) at some point, but I don't
remember what was the conclusion for the future.

(And I don't think Stephen uses Cygwin.  But I may be mistaken.)



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-10 15:39     ` Eli Zaretskii
@ 2022-07-10 18:32       ` Lynn Winebarger
  2022-07-10 21:54         ` Ken Brown
  0 siblings, 1 reply; 18+ messages in thread
From: Lynn Winebarger @ 2022-07-10 18:32 UTC (permalink / raw)
  To: Eli Zaretskii; +Cc: Ken Brown, Stephen Leake, emacs-devel

I see a discussion under bug #50666 -
https://debbugs.gnu.org/cgi/bugreport.cgi?bug=50666 [ note - only
found by google site search on lists.gnu.org for "Ken Brown" "cygwin"
"native", the bug did not show up when I used debbugs's search for
bugs for package "emacs" with "cygwin" in the subject ].
It seems to have focused on the 32-bit architecture, but I only use
64-bit targets.
I'm updating dozens of packages at a time, although the practical
limit seems to be 20-30 in a pass before I get an error about having
run out of some resource (process ids or file descriptors or something
along those lines, not sure).  In any case, if the eln files are
asynchronously getting rebased before being loaded, perhaps there's a
race condition somewhere?
I can file a bug if merited.

Lynn

On Sun, Jul 10, 2022 at 11:39 AM Eli Zaretskii <eliz@gnu.org> wrote:
>
> > From: Lynn Winebarger <owinebar@gmail.com>
> > Date: Sun, 10 Jul 2022 11:10:27 -0400
> > Cc: Stephen Leake <stephen_leake@stephe-leake.org>, emacs-devel <emacs-devel@gnu.org>
> >
> > If Stephen's building native compilation under cygwin, perhaps he's
> > referring to an error message about not being able to load the ELN
> > file, either in the form of a "fork" failure, or an explicit complaint
> > about the dll base address conflicting with another dll that's already
> > been loaded.  For a 64-bit build, a restart of emacs seems to be
> > enough to cure the issue, although I have also had issues that didn't
> > seem to be resolved without doing a "rebase all".
> > I don't know if this behavior is considered a bug (in either libgccjit
> > or emacs) or just a basic limitation of using the cygwin platform.
> > It is definitely an annoyance when either installing packages or when
> > one of the package's features/modules is autoloaded the first time.
>
> This was discussed with Ken Brown (CC'ed) at some point, but I don't
> remember what was the conclusion for the future.
>
> (And I don't think Stephen uses Cygwin.  But I may be mistaken.)



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-10 18:32       ` Lynn Winebarger
@ 2022-07-10 21:54         ` Ken Brown
  2022-07-11  1:52           ` Ken Brown
  0 siblings, 1 reply; 18+ messages in thread
From: Ken Brown @ 2022-07-10 21:54 UTC (permalink / raw)
  To: Lynn Winebarger, Eli Zaretskii; +Cc: Stephen Leake, emacs-devel, Achim Gratz

On 7/10/2022 2:32 PM, Lynn Winebarger wrote:
> I see a discussion under bug #50666 -
> https://debbugs.gnu.org/cgi/bugreport.cgi?bug=50666 [ note - only
> found by google site search on lists.gnu.org for "Ken Brown" "cygwin"
> "native", the bug did not show up when I used debbugs's search for
> bugs for package "emacs" with "cygwin" in the subject ].
> It seems to have focused on the 32-bit architecture, but I only use
> 64-bit targets.

Native compilation is unusable on 32-bit Cygwin, and this is reflected in the 
configure script.  (See the --with-cygwin32-native-compilation configure option.)

In the 64-bit case, Achim Gratz's autorebase postinstall script takes care of 
rebasing the .eln files on a regular basis, provided the user has set things up 
appropriately.  Instructions can be found in the announcement at

   https://cygwin.com/pipermail/cygwin-announce/2022-April/010529.html

In the 3 months since I sent that announcement, I have not heard from a single 
Cygwin user about rebase issues.  This might simply mean that very few users 
have tried the native compilation release.

I myself use that release daily, and I can only recall one instance in which I 
saw a fork failure and had to exit emacs and rebase.

In summary, I would say that native compilation is usable with very occasional 
minor annoyances on 64-bit Cygwin.  But I doubt if I will ever make it the 
default Cygwin build, simply because I don't want to be inundated with emails 
from people who haven't read the release announcement.

Ken



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-10 21:54         ` Ken Brown
@ 2022-07-11  1:52           ` Ken Brown
  2022-07-11  2:39             ` Lynn Winebarger
  0 siblings, 1 reply; 18+ messages in thread
From: Ken Brown @ 2022-07-11  1:52 UTC (permalink / raw)
  To: Lynn Winebarger, Eli Zaretskii; +Cc: Stephen Leake, emacs-devel, Achim Gratz

On 7/10/2022 5:54 PM, Ken Brown wrote:
> On 7/10/2022 2:32 PM, Lynn Winebarger wrote:
>> I see a discussion under bug #50666 -
>> https://debbugs.gnu.org/cgi/bugreport.cgi?bug=50666 [ note - only
>> found by google site search on lists.gnu.org for "Ken Brown" "cygwin"
>> "native", the bug did not show up when I used debbugs's search for
>> bugs for package "emacs" with "cygwin" in the subject ].
>> It seems to have focused on the 32-bit architecture, but I only use
>> 64-bit targets.
> 
> Native compilation is unusable on 32-bit Cygwin, and this is reflected in the 
> configure script.  (See the --with-cygwin32-native-compilation configure option.)
> 
> In the 64-bit case, Achim Gratz's autorebase postinstall script takes care of 
> rebasing the .eln files on a regular basis, provided the user has set things up 
> appropriately.  Instructions can be found in the announcement at
> 
>    https://cygwin.com/pipermail/cygwin-announce/2022-April/010529.html
> 
> In the 3 months since I sent that announcement, I have not heard from a single 
> Cygwin user about rebase issues.  This might simply mean that very few users 
> have tried the native compilation release.
> 
> I myself use that release daily, and I can only recall one instance in which I 
> saw a fork failure and had to exit emacs and rebase.
> 
> In summary, I would say that native compilation is usable with very occasional 
> minor annoyances on 64-bit Cygwin.  But I doubt if I will ever make it the 
> default Cygwin build, simply because I don't want to be inundated with emails 
> from people who haven't read the release announcement.

Lynn,

Rereading your earlier message about problems during package installation, I see 
I didn't really respond to that.  But it has nothing to do with the present bug, 
so please make a fresh bug report and give full details.  And please follow the 
instructions in the announcement I cited.  If you're working in your own build 
of Cygwin emacs that you haven't installed, you might also have to add its 
native-lisp directory to

   /var/lib/rebase/userpath.d/<username>

Ken



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-11  1:52           ` Ken Brown
@ 2022-07-11  2:39             ` Lynn Winebarger
  2022-07-11 13:37               ` Ken Brown
  0 siblings, 1 reply; 18+ messages in thread
From: Lynn Winebarger @ 2022-07-11  2:39 UTC (permalink / raw)
  To: Ken Brown; +Cc: Eli Zaretskii, Stephen Leake, emacs-devel, Achim Gratz

[-- Attachment #1: Type: text/plain, Size: 3001 bytes --]

On Sun, Jul 10, 2022, 9:53 PM Ken Brown <kbrown@cornell.edu> wrote:

> On 7/10/2022 5:54 PM, Ken Brown wrote:
> > On 7/10/2022 2:32 PM, Lynn Winebarger wrote:
> >> I see a discussion under bug #50666 -
> >> https://debbugs.gnu.org/cgi/bugreport.cgi?bug=50666 [ note - only
> >> found by google site search on lists.gnu.org for "Ken Brown" "cygwin"
> >> "native", the bug did not show up when I used debbugs's search for
> >> bugs for package "emacs" with "cygwin" in the subject ].
> >> It seems to have focused on the 32-bit architecture, but I only use
> >> 64-bit targets.
> >
> > Native compilation is unusable on 32-bit Cygwin, and this is reflected
> in the
> > configure script.  (See the --with-cygwin32-native-compilation configure
> option.)
> >
> > In the 64-bit case, Achim Gratz's autorebase postinstall script takes
> care of
> > rebasing the .eln files on a regular basis, provided the user has set
> things up
> > appropriately.  Instructions can be found in the announcement at
> >
> >    https://cygwin.com/pipermail/cygwin-announce/2022-April/010529.html
> >
> > In the 3 months since I sent that announcement, I have not heard from a
> single
> > Cygwin user about rebase issues.  This might simply mean that very few
> users
> > have tried the native compilation release.
>

Will do, but could you include the details of the announcement in the emacs
source distribution somewhere, as is done for the other variants with
specific instructions?  I'm only incidentally building it in a cygwin
environment - it didn't even occur to me to check the general cygwin
mailing list.

>
> > I myself use that release daily, and I can only recall one instance in
> which I
> > saw a fork failure and had to exit emacs and rebase.
> >
> > In summary, I would say that native compilation is usable with very
> occasional
> > minor annoyances on 64-bit Cygwin.  But I doubt if I will ever make it
> the
> > default Cygwin build, simply because I don't want to be inundated with
> emails
> > from people who haven't read the release announcement.
>
> Lynn,
>
> Rereading your earlier message about problems during package installation,
> I see
> I didn't really respond to that.  But it has nothing to do with the
> present bug,
> so please make a fresh bug report and give full details.  And please
> follow the
> instructions in the announcement I cited.  If you're working in your own
> build
> of Cygwin emacs that you haven't installed, you might also have to add its
> native-lisp directory to
>
>    /var/lib/rebase/userpath.d/<username>



I am doing exactly what you surmised, running a build I haven't installed -
in fact that I built with --prefix=/does/not/exist/ to ensure system
installed site-lisp files will not get injected into the load-path while
doing do.  So I'll try that first.  Although I think the packages
generating the fork failures should be going into the cache in my home
directory.
If there are still issues, did you mean to log a bug with cygwin or emacs?

Lynn

[-- Attachment #2: Type: text/html, Size: 4356 bytes --]

^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-09 22:36 ` Tor Kringeland
  2022-07-10  8:58   ` Philip Kaludercic
@ 2022-07-11 12:54   ` Andrea Corallo
  2022-07-12 11:04     ` Tor Kringeland
  1 sibling, 1 reply; 18+ messages in thread
From: Andrea Corallo @ 2022-07-11 12:54 UTC (permalink / raw)
  To: Tor Kringeland; +Cc: Stephen Leake, emacs-devel

Tor Kringeland <tor.kringeland@ntnu.no> writes:

> Stephen Leake <stephen_leake@stephe-leake.org> writes:
>
>> - There should be a simple way to completely disable native compilation;
>> Emacs should never do anything that I can't control. If setting
>> native-comp-deferred-compilation nil is the right way to do this, it
>> should be documented in the native compilation section of the elisp
>> manual.
>
> Doesn't setting
>
>   (setq native-comp-speed -1)
>
> disable native compilation?

Not "completely".  `native-comp-speed' works at function per function
level.  A function copiled with native-comp-speed -1 will still be
enclosed in the produced .eln file but in bytecode form.  Indeed setting
it globally will have as result the production of the same .eln files
but those will include the bytecode definition of the compiled
functions.

BR

  Andrea



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-11  2:39             ` Lynn Winebarger
@ 2022-07-11 13:37               ` Ken Brown
  2022-07-11 13:56                 ` Eli Zaretskii
  0 siblings, 1 reply; 18+ messages in thread
From: Ken Brown @ 2022-07-11 13:37 UTC (permalink / raw)
  To: Lynn Winebarger; +Cc: Eli Zaretskii, Stephen Leake, emacs-devel, Achim Gratz

On 7/10/2022 10:39 PM, Lynn Winebarger wrote:
> On Sun, Jul 10, 2022, 9:53 PM Ken Brown <kbrown@cornell.edu 
> <mailto:kbrown@cornell.edu>> wrote:
> 
>     On 7/10/2022 5:54 PM, Ken Brown wrote:
>      > Native compilation is unusable on 32-bit Cygwin, and this is reflected in
>     the
>      > configure script.  (See the --with-cygwin32-native-compilation configure
>     option.)
>      >
>      > In the 64-bit case, Achim Gratz's autorebase postinstall script takes
>     care of
>      > rebasing the .eln files on a regular basis, provided the user has set
>     things up
>      > appropriately.  Instructions can be found in the announcement at
>      >
>      > https://cygwin.com/pipermail/cygwin-announce/2022-April/010529.html
>     <https://cygwin.com/pipermail/cygwin-announce/2022-April/010529.html>
>      >
>      > In the 3 months since I sent that announcement, I have not heard from a
>     single
>      > Cygwin user about rebase issues.  This might simply mean that very few users
>      > have tried the native compilation release.
> 
> 
> Will do, but could you include the details of the announcement in the emacs 
> source distribution somewhere, as is done for the other variants with specific 
> instructions?  I'm only incidentally building it in a cygwin environment - it 
> didn't even occur to me to check the general cygwin mailing list.

Yes, I'll add something to etc/PROBLEMS unless you or Eli can suggest a better 
place.

>      >
>      > I myself use that release daily, and I can only recall one instance in
>     which I
>      > saw a fork failure and had to exit emacs and rebase.
>      >
>      > In summary, I would say that native compilation is usable with very
>     occasional
>      > minor annoyances on 64-bit Cygwin.  But I doubt if I will ever make it the
>      > default Cygwin build, simply because I don't want to be inundated with
>     emails
>      > from people who haven't read the release announcement.
> 
>     Lynn,
> 
>     Rereading your earlier message about problems during package installation, I
>     see
>     I didn't really respond to that.  But it has nothing to do with the present
>     bug,
>     so please make a fresh bug report and give full details.  And please follow the
>     instructions in the announcement I cited.  If you're working in your own build
>     of Cygwin emacs that you haven't installed, you might also have to add its
>     native-lisp directory to
> 
>         /var/lib/rebase/userpath.d/<username>
> 
> 
> I am doing exactly what you surmised, running a build I haven't installed - in 
> fact that I built with --prefix=/does/not/exist/ to ensure system installed 
> site-lisp files will not get injected into the load-path while doing do.  So 
> I'll try that first.  Although I think the packages generating the fork failures 
> should be going into the cache in my home directory.

That's true, but Cygwin's autorebase facility needs to know about *all* .eln 
files that might be used, in order to try to maintain a conflict-free set of 
base addresses.

And I forgot to say that after you've created 
/var/lib/rebase/userpath.d/<username> with the appropriate paths, you should 
shut down all Cygwin processes and run Cygwin's setup program.  (You don't have 
to install anything; this is just to let the autorebase postinstall script run.)

> If there are still issues, did you mean to log a bug with cygwin or emacs?

I meant emacs.  Thanks.

Ken



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-11 13:37               ` Ken Brown
@ 2022-07-11 13:56                 ` Eli Zaretskii
  2022-07-11 17:19                   ` Ken Brown
  0 siblings, 1 reply; 18+ messages in thread
From: Eli Zaretskii @ 2022-07-11 13:56 UTC (permalink / raw)
  To: Ken Brown; +Cc: owinebar, stephen_leake, emacs-devel, Stromeko

> Date: Mon, 11 Jul 2022 09:37:35 -0400
> Cc: Eli Zaretskii <eliz@gnu.org>,
>  Stephen Leake <stephen_leake@stephe-leake.org>,
>  emacs-devel <emacs-devel@gnu.org>, Achim Gratz <Stromeko@nexgo.de>
> From: Ken Brown <kbrown@cornell.edu>
> 
> > Will do, but could you include the details of the announcement in the emacs 
> > source distribution somewhere, as is done for the other variants with specific 
> > instructions?  I'm only incidentally building it in a cygwin environment - it 
> > didn't even occur to me to check the general cygwin mailing list.
> 
> Yes, I'll add something to etc/PROBLEMS unless you or Eli can suggest a better 
> place.

PROBLEMS is fine for this kind of information.  But please add that to
the release branch, as it is relevant to Emacs 28.

Thanks.



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-11 13:56                 ` Eli Zaretskii
@ 2022-07-11 17:19                   ` Ken Brown
  0 siblings, 0 replies; 18+ messages in thread
From: Ken Brown @ 2022-07-11 17:19 UTC (permalink / raw)
  To: Eli Zaretskii; +Cc: owinebar, stephen_leake, emacs-devel, Stromeko

On 7/11/2022 9:56 AM, Eli Zaretskii wrote:
>> Date: Mon, 11 Jul 2022 09:37:35 -0400
>> From: Ken Brown <kbrown@cornell.edu>
>> Yes, I'll add something to etc/PROBLEMS unless you or Eli can suggest a better
>> place.
> 
> PROBLEMS is fine for this kind of information.  But please add that to
> the release branch, as it is relevant to Emacs 28.

OK, I've done this.  Lynn, I hope this is helpful for you.  If you still have 
problems when installing/updating several packages, we'll have to think harder 
about a fix or workaround.  Or we might just have to add that to etc/PROBLEMS as 
a known limitation of native compilation on Cygwin.

Ken



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-11 12:54   ` Andrea Corallo
@ 2022-07-12 11:04     ` Tor Kringeland
  2022-07-12 11:27       ` Tor Kringeland
  0 siblings, 1 reply; 18+ messages in thread
From: Tor Kringeland @ 2022-07-12 11:04 UTC (permalink / raw)
  To: Andrea Corallo; +Cc: Stephen Leake, emacs-devel

Andrea Corallo <akrl@sdf.org> writes:

> Not "completely".  `native-comp-speed' works at function per function
> level.  A function copiled with native-comp-speed -1 will still be
> enclosed in the produced .eln file but in bytecode form.  Indeed setting
> it globally will have as result the production of the same .eln files
> but those will include the bytecode definition of the compiled
> functions.

I see.  Maybe this could be mentioned explicitly in the manual?  What
the manual says now is consistent with what you say, but when I read it
I got the impression that setting 'native-comp-speed' to -1 in
early-init.el would completely disable native compilation/wouldn't
generate the .eln files at all.


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-12 11:04     ` Tor Kringeland
@ 2022-07-12 11:27       ` Tor Kringeland
  2022-07-12 13:29         ` Eli Zaretskii
  0 siblings, 1 reply; 18+ messages in thread
From: Tor Kringeland @ 2022-07-12 11:27 UTC (permalink / raw)
  To: Andrea Corallo; +Cc: Stephen Leake, emacs-devel

Tor Kringeland <tor.kringeland@ntnu.no> writes:

> I see.  Maybe this could be mentioned explicitly in the manual?  What
> the manual says now is consistent with what you say, but when I read it
> I got the impression that setting 'native-comp-speed' to -1 in
> early-init.el would completely disable native compilation/wouldn't
> generate the .eln files at all.

Nevermind, it seems like Eli Zaretskii thought of the same and already
changed the manual and doc string to be more explicit about this.


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-12 11:27       ` Tor Kringeland
@ 2022-07-12 13:29         ` Eli Zaretskii
  0 siblings, 0 replies; 18+ messages in thread
From: Eli Zaretskii @ 2022-07-12 13:29 UTC (permalink / raw)
  To: Tor Kringeland; +Cc: akrl, stephen_leake, emacs-devel

> From: Tor Kringeland <tor.kringeland@ntnu.no>
> CC: Stephen Leake <stephen_leake@stephe-leake.org>, emacs-devel
>  <emacs-devel@gnu.org>
> Date: Tue, 12 Jul 2022 11:27:29 +0000
> 
> Tor Kringeland <tor.kringeland@ntnu.no> writes:
> 
> > I see.  Maybe this could be mentioned explicitly in the manual?  What
> > the manual says now is consistent with what you say, but when I read it
> > I got the impression that setting 'native-comp-speed' to -1 in
> > early-init.el would completely disable native compilation/wouldn't
> > generate the .eln files at all.
> 
> Nevermind, it seems like Eli Zaretskii thought of the same and already
> changed the manual and doc string to be more explicit about this.

You are welcome ;-)



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: disable automatic native-compilation?
  2022-07-09 16:49 disable automatic native-compilation? Stephen Leake
  2022-07-09 17:23 ` Eli Zaretskii
  2022-07-09 22:36 ` Tor Kringeland
@ 2022-08-06 15:32 ` Lynn Winebarger
  2 siblings, 0 replies; 18+ messages in thread
From: Lynn Winebarger @ 2022-08-06 15:32 UTC (permalink / raw)
  To: Stephen Leake; +Cc: emacs-devel

[-- Attachment #1: Type: text/plain, Size: 240 bytes --]

On Sat, Jul 9, 2022, 12:51 PM Stephen Leake <stephen_leake@stephe-leake.org>
wrote:

> The elisp manual section on native compilation does not say how to
> disable it.
>

I set native-comp-deferred-compilation-deny-list to '("/.*/").

Lynn

[-- Attachment #2: Type: text/html, Size: 829 bytes --]

^ permalink raw reply	[flat|nested] 18+ messages in thread

end of thread, other threads:[~2022-08-06 15:32 UTC | newest]

Thread overview: 18+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-07-09 16:49 disable automatic native-compilation? Stephen Leake
2022-07-09 17:23 ` Eli Zaretskii
2022-07-10 15:10   ` Lynn Winebarger
2022-07-10 15:39     ` Eli Zaretskii
2022-07-10 18:32       ` Lynn Winebarger
2022-07-10 21:54         ` Ken Brown
2022-07-11  1:52           ` Ken Brown
2022-07-11  2:39             ` Lynn Winebarger
2022-07-11 13:37               ` Ken Brown
2022-07-11 13:56                 ` Eli Zaretskii
2022-07-11 17:19                   ` Ken Brown
2022-07-09 22:36 ` Tor Kringeland
2022-07-10  8:58   ` Philip Kaludercic
2022-07-11 12:54   ` Andrea Corallo
2022-07-12 11:04     ` Tor Kringeland
2022-07-12 11:27       ` Tor Kringeland
2022-07-12 13:29         ` Eli Zaretskii
2022-08-06 15:32 ` Lynn Winebarger

Code repositories for project(s) associated with this inbox:

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).