git@vger.kernel.org list mirror (unofficial, one of many)
 help / color / mirror / code / Atom feed
* [PATCH] wrapper: add workaround for open() returning EINTR
@ 2021-02-24  4:43 Jeff King
  2021-02-24  4:46 ` Jeff King
                   ` (2 more replies)
  0 siblings, 3 replies; 15+ messages in thread
From: Jeff King @ 2021-02-24  4:43 UTC (permalink / raw)
  To: git; +Cc: Aleksey Kliger

On some platforms, open() reportedly returns EINTR when opening regular
files and we receive a signal (usually SIGALRM from our progress meter).
This shouldn't happen, as open() should be a restartable syscall, and we
specify SA_RESTART when setting up the alarm handler. So it may actually
be a kernel or libc bug for this to happen. But it has been reported on
at least one version of Linux (on a network filesystem):

  https://lore.kernel.org/git/c8061cce-71e4-17bd-a56a-a5fed93804da@neanderfunk.de/

as well as on macOS starting with Big Sur even on a regular filesystem.

We can work around it by retrying open() calls that get EINTR, just as
we do for read(), etc. Since we don't ever _want_ to interrupt an open()
call, we can get away with just redefining open, rather than insisting
all callsites use xopen().

We actually do have an xopen() wrapper already (and it even does this
retry, though there's no indication of it being an observed problem back
then; it seems simply to have been lifted from xread(), etc). But it is
used hardly anywhere, and isn't suitable for general use because it will
die() on error. In theory we could combine the two, but it's awkward to
do so because of the variable-args interface of open().

The workaround here is enabled all the time, without a Makefile knob,
since it's a complete noop if open() never returns EINTR. I did push it
into its own compat/ source file, though, since it has to #undef our
macro redirection. Putting it in a file with other code risks confusion
if more code is added after that #undef.

Reported-by: Aleksey Kliger <alklig@microsoft.com>
Signed-off-by: Jeff King <peff@peff.net>
---
This was reported to me off-list. Aleksey was kind enough to test the
patch on a system that was reliably showing the problem during the
checkout phase of git-clone (which naturally has both a progress meter
and is calling open() on a ton of files).

I do still think the OS is doing the wrong thing here. But even if I'm
right, it's probably prudent to work around it.

 git-compat-util.h |  4 ++++
 wrapper.c         | 29 +++++++++++++++++++++++++++++
 2 files changed, 33 insertions(+)

diff --git a/git-compat-util.h b/git-compat-util.h
index 838246289c..2be022c422 100644
--- a/git-compat-util.h
+++ b/git-compat-util.h
@@ -788,6 +788,10 @@ int git_vsnprintf(char *str, size_t maxsize,
 		  const char *format, va_list ap);
 #endif
 
+#undef open
+#define open git_open_with_retry
+int git_open_with_retry(const char *path, int flag, ...);
+
 #ifdef __GLIBC_PREREQ
 #if __GLIBC_PREREQ(2, 1)
 #define HAVE_STRCHRNUL
diff --git a/wrapper.c b/wrapper.c
index bcda41e374..130f441064 100644
--- a/wrapper.c
+++ b/wrapper.c
@@ -678,3 +678,32 @@ int is_empty_or_missing_file(const char *filename)
 
 	return !st.st_size;
 }
+
+/*
+ * Do not add other callers of open() after this. Our redefinition
+ * below will ensure the compiler catches any.
+ */
+#undef open
+int git_open_with_retry(const char *path, int flags, ...)
+{
+	mode_t mode = 0;
+	int ret;
+
+	/*
+	 * Also O_TMPFILE would take a mode, but it isn't defined everywhere.
+	 * And anyway, we don't use it in our code base.
+	 */
+	if (flags & O_CREAT) {
+		va_list ap;
+		va_start(ap, flags);
+		mode = va_arg(ap, int);
+		va_end(ap);
+	}
+
+	do {
+		ret = open(path, flags, mode);
+	} while (ret < 0 && errno == EINTR);
+
+	return ret;
+}
+#define open do_not_allow_use_of_bare_open
-- 
2.30.1.1095.g03347429ea

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] wrapper: add workaround for open() returning EINTR
  2021-02-24  4:43 [PATCH] wrapper: add workaround for open() returning EINTR Jeff King
@ 2021-02-24  4:46 ` Jeff King
  2021-02-24  5:36   ` Junio C Hamano
  2021-02-24  4:50 ` [PATCH] wrapper: add workaround for open() returning EINTR Taylor Blau
  2021-02-24  7:20 ` Johannes Sixt
  2 siblings, 1 reply; 15+ messages in thread
From: Jeff King @ 2021-02-24  4:46 UTC (permalink / raw)
  To: git; +Cc: Aleksey Kliger

On Tue, Feb 23, 2021 at 11:43:16PM -0500, Jeff King wrote:

> The workaround here is enabled all the time, without a Makefile knob,
> since it's a complete noop if open() never returns EINTR. I did push it
> into its own compat/ source file, though, since it has to #undef our
> macro redirection. Putting it in a file with other code risks confusion
> if more code is added after that #undef.

Whoops, sorry, I had a last-minute change of heart here and just stuck
it in wrapper.c with a bit of preprocessor magic to guard it. It felt
weird having compat/open.c, when the rest of compat/ is always
conditional on Makefile knobs. But I'm happy to go the other way if
anybody feels strongly.

-Peff

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] wrapper: add workaround for open() returning EINTR
  2021-02-24  4:43 [PATCH] wrapper: add workaround for open() returning EINTR Jeff King
  2021-02-24  4:46 ` Jeff King
@ 2021-02-24  4:50 ` Taylor Blau
  2021-02-24  7:20 ` Johannes Sixt
  2 siblings, 0 replies; 15+ messages in thread
From: Taylor Blau @ 2021-02-24  4:50 UTC (permalink / raw)
  To: Jeff King; +Cc: git, Aleksey Kliger

On Tue, Feb 23, 2021 at 11:43:15PM -0500, Jeff King wrote:
> The workaround here is enabled all the time, without a Makefile knob,
> since it's a complete noop if open() never returns EINTR. I did push it
> into its own compat/ source file, though, since it has to #undef our
> macro redirection. Putting it in a file with other code risks confusion
> if more code is added after that #undef.

Hmm. The patch below defines it in wrapper.c. Intentional?

> I do still think the OS is doing the wrong thing here. But even if I'm
> right, it's probably prudent to work around it.

Regardless of the above, I agree that if your explanation is true (and I
have no reason to believe that it isn't) that the OS is indeed doing the
wrong thing here.

That patch below looks quite reasonable, thanks.


Thanks,
Taylor

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] wrapper: add workaround for open() returning EINTR
  2021-02-24  4:46 ` Jeff King
@ 2021-02-24  5:36   ` Junio C Hamano
  2021-02-24 18:20     ` Jeff King
  0 siblings, 1 reply; 15+ messages in thread
From: Junio C Hamano @ 2021-02-24  5:36 UTC (permalink / raw)
  To: Jeff King; +Cc: git, Aleksey Kliger

Jeff King <peff@peff.net> writes:

> On Tue, Feb 23, 2021 at 11:43:16PM -0500, Jeff King wrote:
>
>> The workaround here is enabled all the time, without a Makefile knob,
>> since it's a complete noop if open() never returns EINTR. I did push it
>> into its own compat/ source file, though, since it has to #undef our
>> macro redirection. Putting it in a file with other code risks confusion
>> if more code is added after that #undef.
>
> Whoops, sorry, I had a last-minute change of heart here and just stuck
> it in wrapper.c with a bit of preprocessor magic to guard it. It felt
> weird having compat/open.c, when the rest of compat/ is always
> conditional on Makefile knobs. But I'm happy to go the other way if
> anybody feels strongly.

Hmph, just like other workarounds, shouldn't this be "if your
platform screws this up, define this knob to work it around"?  That
would make it fit better with the rest of compat/.

I dunno.  A no-op wrapper makes the code less transparent, which I
am somewhat unhappy.  But a wrapper that is always used even on
platforms that do not need might give us a better chance of noticing
a bug in the wrapper itself, making it less likely for us to leave
only the users of minority broken platforms behind.  So...



^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] wrapper: add workaround for open() returning EINTR
  2021-02-24  4:43 [PATCH] wrapper: add workaround for open() returning EINTR Jeff King
  2021-02-24  4:46 ` Jeff King
  2021-02-24  4:50 ` [PATCH] wrapper: add workaround for open() returning EINTR Taylor Blau
@ 2021-02-24  7:20 ` Johannes Sixt
  2021-02-24 18:23   ` Jeff King
  2 siblings, 1 reply; 15+ messages in thread
From: Johannes Sixt @ 2021-02-24  7:20 UTC (permalink / raw)
  To: Jeff King, git; +Cc: Aleksey Kliger

Am 24.02.21 um 05:43 schrieb Jeff King:
> The workaround here is enabled all the time, without a Makefile knob,
> since it's a complete noop if open() never returns EINTR. I did push it
> into its own compat/ source file, though, since it has to #undef our
> macro redirection. Putting it in a file with other code risks confusion
> if more code is added after that #undef.

I'm not so much opposed to "enable it all the time" in general, but when
we already have an override of open(), like for the Windows case in
compat/mingw.h, I find it a bit rough to put another wrapper around it,
even more so since we won't have the EINTR problem on Windows due to the
absence of signals.

-- Hannes

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] wrapper: add workaround for open() returning EINTR
  2021-02-24  5:36   ` Junio C Hamano
@ 2021-02-24 18:20     ` Jeff King
  2021-02-26  6:14       ` [PATCH v2] Makefile: add OPEN_RETURNS_EINTR knob Jeff King
  0 siblings, 1 reply; 15+ messages in thread
From: Jeff King @ 2021-02-24 18:20 UTC (permalink / raw)
  To: Junio C Hamano; +Cc: git, Aleksey Kliger

On Tue, Feb 23, 2021 at 09:36:25PM -0800, Junio C Hamano wrote:

> Jeff King <peff@peff.net> writes:
> 
> > On Tue, Feb 23, 2021 at 11:43:16PM -0500, Jeff King wrote:
> >
> >> The workaround here is enabled all the time, without a Makefile knob,
> >> since it's a complete noop if open() never returns EINTR. I did push it
> >> into its own compat/ source file, though, since it has to #undef our
> >> macro redirection. Putting it in a file with other code risks confusion
> >> if more code is added after that #undef.
> >
> > Whoops, sorry, I had a last-minute change of heart here and just stuck
> > it in wrapper.c with a bit of preprocessor magic to guard it. It felt
> > weird having compat/open.c, when the rest of compat/ is always
> > conditional on Makefile knobs. But I'm happy to go the other way if
> > anybody feels strongly.
> 
> Hmph, just like other workarounds, shouldn't this be "if your
> platform screws this up, define this knob to work it around"?  That
> would make it fit better with the rest of compat/.

I actually started that way, but then I got stuck deciding which
platforms in config.mak.uname to enable it for. We've seen it on Linux
and on macOS. I don't know how prevalent it is on those platforms (or
whether it is indeed a bug that may even get fixed). Since the check is
basically zero-cost at run-time, it seemed like it was worth just being
on guard for it so that nobody ever had to worry about it again.

But I don't feel that strongly. I'd be happy to do it with a Makefile
knob if you prefer.

> I dunno.  A no-op wrapper makes the code less transparent, which I
> am somewhat unhappy.

Yeah. I think the run-time cost is essentially zero, but it does add a
bit of complexity if you are debugging. And handling the magic mode
vararg definitely makes it uglier.

The other thing that gives me pause is that there may be other syscalls
which need the same treatment. Here's a recent discussion that Big Sur
may also return EINTR from the opendir() family:

  https://github.com/dotnet/runtime/issues/47584

Git is presumably susceptible to that, too; it's just that we call
open() a lot more frequently, so that was the obvious one that came up.

I don't think there is a more general way to address this, though. We'd
have to wrap each suspected syscall (but we could presumably at least
tie them all to a single Makefile knob). I'm not excited at the prospect
of preemptively adding such wrappers when we aren't even 100% sure that
there is a problem.

> But a wrapper that is always used even on
> platforms that do not need might give us a better chance of noticing
> a bug in the wrapper itself, making it less likely for us to leave
> only the users of minority broken platforms behind.  So...

That part I'm less worried about. It sounds like we'd want it on for
recent versions of macOS, which would give us pretty wide coverage to
notice a bug.

-Peff

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] wrapper: add workaround for open() returning EINTR
  2021-02-24  7:20 ` Johannes Sixt
@ 2021-02-24 18:23   ` Jeff King
  2021-02-26  6:17     ` Jeff King
  0 siblings, 1 reply; 15+ messages in thread
From: Jeff King @ 2021-02-24 18:23 UTC (permalink / raw)
  To: Johannes Sixt; +Cc: git, Aleksey Kliger

On Wed, Feb 24, 2021 at 08:20:57AM +0100, Johannes Sixt wrote:

> Am 24.02.21 um 05:43 schrieb Jeff King:
> > The workaround here is enabled all the time, without a Makefile knob,
> > since it's a complete noop if open() never returns EINTR. I did push it
> > into its own compat/ source file, though, since it has to #undef our
> > macro redirection. Putting it in a file with other code risks confusion
> > if more code is added after that #undef.
> 
> I'm not so much opposed to "enable it all the time" in general, but when
> we already have an override of open(), like for the Windows case in
> compat/mingw.h, I find it a bit rough to put another wrapper around it,
> even more so since we won't have the EINTR problem on Windows due to the
> absence of signals.

That's fair. I don't think my new wrapper would interact well with
mingw_open(). They are both trying to #define open. I think since mine
comes later in git-compat-util.h, it is probably overriding mingw_open()
completely on Windows, which is quite bad (we call into my function in
wrapper.c, but then its "#undef open" means we get the original open(),
not the previously defined wrapper).

-Peff

^ permalink raw reply	[flat|nested] 15+ messages in thread

* [PATCH v2] Makefile: add OPEN_RETURNS_EINTR knob
  2021-02-24 18:20     ` Jeff King
@ 2021-02-26  6:14       ` Jeff King
  2021-02-26 22:17         ` Junio C Hamano
  0 siblings, 1 reply; 15+ messages in thread
From: Jeff King @ 2021-02-26  6:14 UTC (permalink / raw)
  To: Junio C Hamano; +Cc: git, Aleksey Kliger

On Wed, Feb 24, 2021 at 01:20:44PM -0500, Jeff King wrote:

> > Hmph, just like other workarounds, shouldn't this be "if your
> > platform screws this up, define this knob to work it around"?  That
> > would make it fit better with the rest of compat/.
> 
> I actually started that way, but then I got stuck deciding which
> platforms in config.mak.uname to enable it for. We've seen it on Linux
> and on macOS. I don't know how prevalent it is on those platforms (or
> whether it is indeed a bug that may even get fixed). Since the check is
> basically zero-cost at run-time, it seemed like it was worth just being
> on guard for it so that nobody ever had to worry about it again.
> 
> But I don't feel that strongly. I'd be happy to do it with a Makefile
> knob if you prefer.

So here's a v2 that turns it into a Makefile knob.

This was motivated by two things:

  - it must not be enabled on Windows, where it overrides the existing
    mingw_open()

  - I couldn't replicate it on a Big Sur machine, using the same
    sequence that Aleksey did (basically just cloning and checking out
    dotnet/runtime.git from scratch).

    So it's not entirely clear to me how widespread the problem is. A
    conservative approach (at least in terms of change from the status
    quo) is to introduce this knob but not enable it automatically
    (which this patch does). And then if somebody runs into it, they can
    try turning the knob.

    It's less conservative in the sense that we won't be preemptively
    protecting people. I dunno. Once we have the knob, it would be
    pretty easy to just flip the default.

-- >8 --
Subject: [PATCH] Makefile: add OPEN_RETURNS_EINTR knob

On some platforms, open() reportedly returns EINTR when opening regular
files and we receive a signal (usually SIGALRM from our progress meter).
This shouldn't happen, as open() should be a restartable syscall, and we
specify SA_RESTART when setting up the alarm handler. So it may actually
be a kernel or libc bug for this to happen. But it has been reported on
at least one version of Linux (on a network filesystem):

  https://lore.kernel.org/git/c8061cce-71e4-17bd-a56a-a5fed93804da@neanderfunk.de/

as well as on macOS starting with Big Sur even on a regular filesystem.

We can work around it by retrying open() calls that get EINTR, just as
we do for read(), etc. Since we don't ever _want_ to interrupt an open()
call, we can get away with just redefining open, rather than insisting
all callsites use xopen().

We actually do have an xopen() wrapper already (and it even does this
retry, though there's no indication of it being an observed problem back
then; it seems simply to have been lifted from xread(), etc). But it is
used hardly anywhere, and isn't suitable for general use because it will
die() on error. In theory we could combine the two, but it's awkward to
do so because of the variable-args interface of open().

This patch adds a Makefile knob for enabling the workaround. It's not
enabled by default for any platforms in config.mak.uname yet, as we
don't have enough data to decide how common this is (I have not been
able to reproduce on either Linux or Big Sur myself). It may be worth
enabling preemptively anyway, since the cost is pretty low (if we don't
see an EINTR, it's just an extra conditional).

However, note that we must not enable this on Windows. It doesn't do
anything there, and the macro overrides the existing mingw_open()
redirection. I've added a preemptive #undef here in the mingw header
(which is processed first) to just quietly disable it (we could also
make it an #error, but there is little point in being so aggressive).

Reported-by: Aleksey Kliger <alklig@microsoft.com>
Signed-off-by: Jeff King <peff@peff.net>
---
 Makefile          |  7 +++++++
 compat/mingw.h    |  1 +
 compat/open.c     | 25 +++++++++++++++++++++++++
 git-compat-util.h |  6 ++++++
 4 files changed, 39 insertions(+)
 create mode 100644 compat/open.c

diff --git a/Makefile b/Makefile
index 9b1bde2e0e..c4872ca16e 100644
--- a/Makefile
+++ b/Makefile
@@ -22,6 +22,9 @@ all::
 # when attempting to read from an fopen'ed directory (or even to fopen
 # it at all).
 #
+# Define OPEN_RETURNS_EINTR if your open() system call may return EINTR
+# when a signal is received (as opposed to restarting).
+#
 # Define NO_OPENSSL environment variable if you do not have OpenSSL.
 #
 # Define USE_LIBPCRE if you have and want to use libpcre. Various
@@ -1538,6 +1541,10 @@ ifdef FREAD_READS_DIRECTORIES
 	COMPAT_CFLAGS += -DFREAD_READS_DIRECTORIES
 	COMPAT_OBJS += compat/fopen.o
 endif
+ifdef OPEN_RETURNS_EINTR
+	COMPAT_CFLAGS += -DOPEN_RETURNS_EINTR
+	COMPAT_OBJS += compat/open.o
+endif
 ifdef NO_SYMLINK_HEAD
 	BASIC_CFLAGS += -DNO_SYMLINK_HEAD
 endif
diff --git a/compat/mingw.h b/compat/mingw.h
index af8eddd73e..c9a52ad64a 100644
--- a/compat/mingw.h
+++ b/compat/mingw.h
@@ -227,6 +227,7 @@ int mingw_rmdir(const char *path);
 
 int mingw_open (const char *filename, int oflags, ...);
 #define open mingw_open
+#undef OPEN_RETURNS_EINTR
 
 int mingw_fgetc(FILE *stream);
 #define fgetc mingw_fgetc
diff --git a/compat/open.c b/compat/open.c
new file mode 100644
index 0000000000..eb3754a23b
--- /dev/null
+++ b/compat/open.c
@@ -0,0 +1,25 @@
+#include "git-compat-util.h"
+
+#undef open
+int git_open_with_retry(const char *path, int flags, ...)
+{
+	mode_t mode = 0;
+	int ret;
+
+	/*
+	 * Also O_TMPFILE would take a mode, but it isn't defined everywhere.
+	 * And anyway, we don't use it in our code base.
+	 */
+	if (flags & O_CREAT) {
+		va_list ap;
+		va_start(ap, flags);
+		mode = va_arg(ap, int);
+		va_end(ap);
+	}
+
+	do {
+		ret = open(path, flags, mode);
+	} while (ret < 0 && errno == EINTR);
+
+	return ret;
+}
diff --git a/git-compat-util.h b/git-compat-util.h
index 838246289c..551cc9f22f 100644
--- a/git-compat-util.h
+++ b/git-compat-util.h
@@ -788,6 +788,12 @@ int git_vsnprintf(char *str, size_t maxsize,
 		  const char *format, va_list ap);
 #endif
 
+#ifdef OPEN_RETURNS_EINTR
+#undef open
+#define open git_open_with_retry
+int git_open_with_retry(const char *path, int flag, ...);
+#endif
+
 #ifdef __GLIBC_PREREQ
 #if __GLIBC_PREREQ(2, 1)
 #define HAVE_STRCHRNUL
-- 
2.31.0.rc0.520.g3ffb8d01f4


^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] wrapper: add workaround for open() returning EINTR
  2021-02-24 18:23   ` Jeff King
@ 2021-02-26  6:17     ` Jeff King
  0 siblings, 0 replies; 15+ messages in thread
From: Jeff King @ 2021-02-26  6:17 UTC (permalink / raw)
  To: Johannes Sixt; +Cc: git, Aleksey Kliger

On Wed, Feb 24, 2021 at 01:23:51PM -0500, Jeff King wrote:

> > I'm not so much opposed to "enable it all the time" in general, but when
> > we already have an override of open(), like for the Windows case in
> > compat/mingw.h, I find it a bit rough to put another wrapper around it,
> > even more so since we won't have the EINTR problem on Windows due to the
> > absence of signals.
> 
> That's fair. I don't think my new wrapper would interact well with
> mingw_open(). They are both trying to #define open. I think since mine
> comes later in git-compat-util.h, it is probably overriding mingw_open()
> completely on Windows, which is quite bad (we call into my function in
> wrapper.c, but then its "#undef open" means we get the original open(),
> not the previously defined wrapper).

Thanks again for pointing out mingw_open(). The v2 I posted should avoid
any bad interactions there, even if we do end up enabling it all the
time.

By the way, I did notice something funny when looking at mingw_open().
It unconditionally does:

          va_start(args, oflags);
          mode = va_arg(args, int);
          va_end(args);

no matter what we see in oflags. But callers who are not passing O_CREAT
will not provide a third argument at all. So probably it is loading
random trash from the stack and passing it around in the "mode"
variable. This doesn't hurt anything, because ultimately we pass the
trash along to the real open function, which will ignore it without
O_CREAT. But it is definitely undefined behavior to call va_arg() at all
in this instance.

-Peff

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH v2] Makefile: add OPEN_RETURNS_EINTR knob
  2021-02-26  6:14       ` [PATCH v2] Makefile: add OPEN_RETURNS_EINTR knob Jeff King
@ 2021-02-26 22:17         ` Junio C Hamano
  2021-03-01  9:29           ` [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur Jeff King
  0 siblings, 1 reply; 15+ messages in thread
From: Junio C Hamano @ 2021-02-26 22:17 UTC (permalink / raw)
  To: Jeff King; +Cc: git, Aleksey Kliger

Jeff King <peff@peff.net> writes:

> However, note that we must not enable this on Windows. It doesn't do
> anything there, and the macro overrides the existing mingw_open()
> redirection. I've added a preemptive #undef here in the mingw header
> (which is processed first) to just quietly disable it (we could also
> make it an #error, but there is little point in being so aggressive).
> ...
> diff --git a/compat/open.c b/compat/open.c
> new file mode 100644
> index 0000000000..eb3754a23b
> --- /dev/null
> +++ b/compat/open.c
> @@ -0,0 +1,25 @@
> +#include "git-compat-util.h"
> +
> +#undef open
> +int git_open_with_retry(const char *path, int flags, ...)
> +{
> +	mode_t mode = 0;
> +	int ret;
> +
> +	/*
> +	 * Also O_TMPFILE would take a mode, but it isn't defined everywhere.
> +	 * And anyway, we don't use it in our code base.
> +	 */

That is being extra careful---I like it very much.

> +	if (flags & O_CREAT) {
> +		va_list ap;
> +		va_start(ap, flags);
> +		mode = va_arg(ap, int);
> +		va_end(ap);
> +	}
> +
> +	do {
> +		ret = open(path, flags, mode);
> +	} while (ret < 0 && errno == EINTR);
> +
> +	return ret;
> +}

Thanks.

> diff --git a/git-compat-util.h b/git-compat-util.h
> index 838246289c..551cc9f22f 100644
> --- a/git-compat-util.h
> +++ b/git-compat-util.h
> @@ -788,6 +788,12 @@ int git_vsnprintf(char *str, size_t maxsize,
>  		  const char *format, va_list ap);
>  #endif
>  
> +#ifdef OPEN_RETURNS_EINTR
> +#undef open
> +#define open git_open_with_retry
> +int git_open_with_retry(const char *path, int flag, ...);
> +#endif
> +
>  #ifdef __GLIBC_PREREQ
>  #if __GLIBC_PREREQ(2, 1)
>  #define HAVE_STRCHRNUL

^ permalink raw reply	[flat|nested] 15+ messages in thread

* [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur
  2021-02-26 22:17         ` Junio C Hamano
@ 2021-03-01  9:29           ` Jeff King
  2021-03-01 17:17             ` Junio C Hamano
  2021-03-01 23:57             ` Junio C Hamano
  0 siblings, 2 replies; 15+ messages in thread
From: Jeff King @ 2021-03-01  9:29 UTC (permalink / raw)
  To: Junio C Hamano; +Cc: git, Aleksey Kliger

On Fri, Feb 26, 2021 at 02:17:53PM -0800, Junio C Hamano wrote:

> > +	/*
> > +	 * Also O_TMPFILE would take a mode, but it isn't defined everywhere.
> > +	 * And anyway, we don't use it in our code base.
> > +	 */
> 
> That is being extra careful---I like it very much.

I wondered what would happen if my "anyway" above is wrong. We at least
would not invoke undefined behavior (because we'd avoid looking at the
mode parameter even though it exists), but would pass a "0" mode to the
real open(). Presumably somebody would notice that. :)

> > +	if (flags & O_CREAT) {
> > +		va_list ap;
> > +		va_start(ap, flags);
> > +		mode = va_arg(ap, int);
> > +		va_end(ap);
> > +	}
> > +
> > +	do {
> > +		ret = open(path, flags, mode);
> > +	} while (ret < 0 && errno == EINTR);
> > +
> > +	return ret;
> > +}
> 
> Thanks.

I got another off-list report of the problem. I think we probably want
to do this on top:

-- >8 --
Subject: config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur

We've had mixed reports on whether the latest release of macOS needs
this Makefile knob set. In most reported cases, there's antivirus
software running (which one might imagine could cause an open() call to
be delayed). However, one of the (off-list) reports I've gotten
indicated that it happened on an otherwise clean install of Big Sur.

Since the symptom is so bad (checkout randomly fails to write several
fails when the progress meter kicks in), and since the workaround is so
lightweight (if we don't see EINTR, it's just an extra conditional
check), let's just turn it on by default.

Signed-off-by: Jeff King <peff@peff.net>
---
Apparently Big Sur jumped from macOS 10.x to 11.x. But our "uname -r"
check gives the "Darwin version", in which it is 20.x (following 19.x
for the previous version). At least according to some sources I found
online. :) So that is good, because otherwise all of our uname_R checks
here would have been broken. I don't have a Big Sur machine handy to
test with, but I believe this should do what we want.

 config.mak.uname | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/config.mak.uname b/config.mak.uname
index e22d4b6d67..d204c20a64 100644
--- a/config.mak.uname
+++ b/config.mak.uname
@@ -124,6 +124,9 @@ ifeq ($(uname_S),Darwin)
 	ifeq ($(shell test "`expr "$(uname_R)" : '\([0-9][0-9]*\)\.'`" -ge 11 && echo 1),1)
 		HAVE_GETDELIM = YesPlease
 	endif
+	ifeq ($(shell test "`expr "$(uname_R)" : '\([0-9][0-9]*\)\.'`" -ge 20 && echo 1),1)
+		OPEN_RETURNS_EINTR = UnfortunatelyYes
+	endif
 	NO_MEMMEM = YesPlease
 	USE_ST_TIMESPEC = YesPlease
 	HAVE_DEV_TTY = YesPlease
-- 
2.31.0.rc0.521.g56be7fa5e1


^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur
  2021-03-01  9:29           ` [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur Jeff King
@ 2021-03-01 17:17             ` Junio C Hamano
  2021-03-01 23:57             ` Junio C Hamano
  1 sibling, 0 replies; 15+ messages in thread
From: Junio C Hamano @ 2021-03-01 17:17 UTC (permalink / raw)
  To: Jeff King; +Cc: git, Aleksey Kliger

Jeff King <peff@peff.net> writes:

> I got another off-list report of the problem. I think we probably want
> to do this on top:
>
> -- >8 --
> Subject: config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur
>
> We've had mixed reports on whether the latest release of macOS needs
> this Makefile knob set. In most reported cases, there's antivirus
> software running (which one might imagine could cause an open() call to
> be delayed). However, one of the (off-list) reports I've gotten
> indicated that it happened on an otherwise clean install of Big Sur.
>
> Since the symptom is so bad (checkout randomly fails to write several
> fails when the progress meter kicks in), and since the workaround is so
> lightweight (if we don't see EINTR, it's just an extra conditional
> check), let's just turn it on by default.
>
> Signed-off-by: Jeff King <peff@peff.net>
> ---
> Apparently Big Sur jumped from macOS 10.x to 11.x. But our "uname -r"
> check gives the "Darwin version", in which it is 20.x (following 19.x
> for the previous version). At least according to some sources I found
> online. :) So that is good, because otherwise all of our uname_R checks
> here would have been broken. I don't have a Big Sur machine handy to
> test with, but I believe this should do what we want.

Thanks.

>  config.mak.uname | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/config.mak.uname b/config.mak.uname
> index e22d4b6d67..d204c20a64 100644
> --- a/config.mak.uname
> +++ b/config.mak.uname
> @@ -124,6 +124,9 @@ ifeq ($(uname_S),Darwin)
>  	ifeq ($(shell test "`expr "$(uname_R)" : '\([0-9][0-9]*\)\.'`" -ge 11 && echo 1),1)
>  		HAVE_GETDELIM = YesPlease
>  	endif
> +	ifeq ($(shell test "`expr "$(uname_R)" : '\([0-9][0-9]*\)\.'`" -ge 20 && echo 1),1)
> +		OPEN_RETURNS_EINTR = UnfortunatelyYes
> +	endif
>  	NO_MEMMEM = YesPlease
>  	USE_ST_TIMESPEC = YesPlease
>  	HAVE_DEV_TTY = YesPlease

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur
  2021-03-01  9:29           ` [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur Jeff King
  2021-03-01 17:17             ` Junio C Hamano
@ 2021-03-01 23:57             ` Junio C Hamano
  2021-03-03 13:41               ` Jeff King
  1 sibling, 1 reply; 15+ messages in thread
From: Junio C Hamano @ 2021-03-01 23:57 UTC (permalink / raw)
  To: Jeff King, Joachim Schmitz, Matt Kraai, Randall S. Becker
  Cc: git, Aleksey Kliger

Jeff King <peff@peff.net> writes:

> I got another off-list report of the problem. I think we probably want
> to do this on top:

Queued and pushed out.

I wonder if these hits for SA_RESTART in config.mak.uname would be a
good way to guide us.

    [6c109904 (Port to HP NonStop, 2012-09-19)]
            # Not detected (nor checked for) by './configure'.
            # We don't have SA_RESTART on NonStop, unfortunalety.
            COMPAT_CFLAGS += -DSA_RESTART=0

    [40036bed (Port to QNX, 2012-12-18)]
    ifeq ($(uname_S),QNX)
            COMPAT_CFLAGS += -DSA_RESTART=0

One caveat is that we do not know if their system headers hide the
real implementation of open(2) behind a C preprocessor macro, in
whcih case OPEN_RETURNS_EINTR wrapper may not work correctly.

>  config.mak.uname | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/config.mak.uname b/config.mak.uname
> index e22d4b6d67..d204c20a64 100644
> --- a/config.mak.uname
> +++ b/config.mak.uname
> @@ -124,6 +124,9 @@ ifeq ($(uname_S),Darwin)
>  	ifeq ($(shell test "`expr "$(uname_R)" : '\([0-9][0-9]*\)\.'`" -ge 11 && echo 1),1)
>  		HAVE_GETDELIM = YesPlease
>  	endif
> +	ifeq ($(shell test "`expr "$(uname_R)" : '\([0-9][0-9]*\)\.'`" -ge 20 && echo 1),1)
> +		OPEN_RETURNS_EINTR = UnfortunatelyYes
> +	endif
>  	NO_MEMMEM = YesPlease
>  	USE_ST_TIMESPEC = YesPlease
>  	HAVE_DEV_TTY = YesPlease

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur
  2021-03-01 23:57             ` Junio C Hamano
@ 2021-03-03 13:41               ` Jeff King
  2021-03-04  0:47                 ` Junio C Hamano
  0 siblings, 1 reply; 15+ messages in thread
From: Jeff King @ 2021-03-03 13:41 UTC (permalink / raw)
  To: Junio C Hamano
  Cc: Joachim Schmitz, Matt Kraai, Randall S. Becker, git, Aleksey Kliger

On Mon, Mar 01, 2021 at 03:57:35PM -0800, Junio C Hamano wrote:

> Jeff King <peff@peff.net> writes:
> 
> > I got another off-list report of the problem. I think we probably want
> > to do this on top:
> 
> Queued and pushed out.

Thanks. I guess we're a bit late to make it into the upcoming release.
Certainly we have survived for many years without this particular
bugfix, so in that sense it is not urgent. But I do wonder if we will
see more reports as more people start using the new macOS release. So it
might be good to keep in mind for maint, if we cut a minor release.

Or alternatively, we could include _just_ the first patch. That's low
risk, since you have to enable to knob yourself, but it gives people an
option if they run into the symptom. But even that is probably not that
urgent. People can also cherry-pick the patch, after all (and a
distributor like homebrew can probably include the patch in their recipe
if need be).

> I wonder if these hits for SA_RESTART in config.mak.uname would be a
> good way to guide us.
> 
>     [6c109904 (Port to HP NonStop, 2012-09-19)]
>             # Not detected (nor checked for) by './configure'.
>             # We don't have SA_RESTART on NonStop, unfortunalety.
>             COMPAT_CFLAGS += -DSA_RESTART=0
> 
>     [40036bed (Port to QNX, 2012-12-18)]
>     ifeq ($(uname_S),QNX)
>             COMPAT_CFLAGS += -DSA_RESTART=0

I'm inclined to leave them alone until somebody with access to such a
system can look further into it. After all, if you do not have
SA_RESTART, you might not even have EINTR in the first place.

> One caveat is that we do not know if their system headers hide the
> real implementation of open(2) behind a C preprocessor macro, in
> whcih case OPEN_RETURNS_EINTR wrapper may not work correctly.

Yeah. I didn't think about that when I did the original "just do it
everywhere" patch. But that is exactly what caused the problem on
Windows (not a system macro, but in fact our own!). So I'm glad to have
backed it off to a Makefile knob.

-Peff

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur
  2021-03-03 13:41               ` Jeff King
@ 2021-03-04  0:47                 ` Junio C Hamano
  0 siblings, 0 replies; 15+ messages in thread
From: Junio C Hamano @ 2021-03-04  0:47 UTC (permalink / raw)
  To: Jeff King
  Cc: Joachim Schmitz, Matt Kraai, Randall S. Becker, git, Aleksey Kliger

Jeff King <peff@peff.net> writes:

> Thanks. I guess we're a bit late to make it into the upcoming release.
> Certainly we have survived for many years without this particular
> bugfix, so in that sense it is not urgent. But I do wonder if we will
> see more reports as more people start using the new macOS release. So it
> might be good to keep in mind for maint, if we cut a minor release.
>
> Or alternatively, we could include _just_ the first patch. That's low
> risk, since you have to enable to knob yourself, but it gives people an
> option if they run into the symptom. But even that is probably not that
> urgent. People can also cherry-pick the patch, after all (and a
> distributor like homebrew can probably include the patch in their recipe
> if need be).

True.  The topic is forked at somewhere mergeable to 'maint' so
distro folks who are interested can merge them down later.  I think
it is small and safe enough to merge them both down to the upcoming
release, though.


^ permalink raw reply	[flat|nested] 15+ messages in thread

end of thread, other threads:[~2021-03-04  1:11 UTC | newest]

Thread overview: 15+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-02-24  4:43 [PATCH] wrapper: add workaround for open() returning EINTR Jeff King
2021-02-24  4:46 ` Jeff King
2021-02-24  5:36   ` Junio C Hamano
2021-02-24 18:20     ` Jeff King
2021-02-26  6:14       ` [PATCH v2] Makefile: add OPEN_RETURNS_EINTR knob Jeff King
2021-02-26 22:17         ` Junio C Hamano
2021-03-01  9:29           ` [PATCH] config.mak.uname: enable OPEN_RETURNS_EINTR for macOS Big Sur Jeff King
2021-03-01 17:17             ` Junio C Hamano
2021-03-01 23:57             ` Junio C Hamano
2021-03-03 13:41               ` Jeff King
2021-03-04  0:47                 ` Junio C Hamano
2021-02-24  4:50 ` [PATCH] wrapper: add workaround for open() returning EINTR Taylor Blau
2021-02-24  7:20 ` Johannes Sixt
2021-02-24 18:23   ` Jeff King
2021-02-26  6:17     ` Jeff King

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

	https://80x24.org/mirrors/git.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).