All of lore.kernel.org
 help / color / mirror / Atom feed
From: Yann E. MORIN <yann.morin.1998@free.fr>
To: buildroot@busybox.net
Subject: [Buildroot] [PATCHv2] package/gcc/9.3.0: fix host-gcc-final when ccache is used
Date: Thu, 21 May 2020 14:12:28 +0200	[thread overview]
Message-ID: <20200521121228.GM27030@scaer> (raw)
In-Reply-To: <20200520224543.2345022-1-romain.naour@gmail.com>

Romain., All,

On 2020-05-21 00:45 +0200, Romain Naour spake thusly:
> As remported by several Buildroot users [1][2][3], the gcc build

*reported

> fail while running selftests makefile target.

*fails

Otherwise, see below, I think I have a better explanation and proposal
for a fix...

> The problem only occurs when ccache is used with gcc 9 and 10,
> probably due to a race condition.
> 
> While debuging with "make -p" we can notice that s-selftest-c target
> contain only "cc1" as dependency instead of cc1 and SELFTEST_DEPS [4].
> 
>   s-selftest-c: cc1
> 
> While the build is failing, the s-selftest-c dependencies recipe is
> still running and reported as a bug by make.
> 
>   "Dependencies recipe running (THIS IS A BUG)."
> 
> A change [5] in gcc 9 seems to introduce the problem since we can't
> reproduce this problem with gcc 8.
> 
> For now add explicitely xgcc dependency to s-selftest-c and
> s-selftest-c++. It not clear why SELFTEST_DEPS is lost.
> 
> [1] http://lists.busybox.net/pipermail/buildroot/2020-May/282171.html
> [2] http://lists.busybox.net/pipermail/buildroot/2020-May/282766.html
> [3] https://github.com/cirosantilli/linux-kernel-module-cheat/issues/108
> [4] https://gcc.gnu.org/git/?p=gcc.git;a=blob;f=gcc/c/Make-lang.in;h=bfae6fd2549c4f728816cd355fa9739dcc08fcde;hb=033eb5671769a4c681a44aad08a454e667e08502#l120
> [5] https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=033eb5671769a4c681a44aad08a454e667e08502
> 
> Signed-off-by: Romain Naour <romain.naour@gmail.com>
> Cc: Ben Dakin-Norris <ben.dakin-norris@navtechradar.com>
> Cc: Maxim Kochetkov <fido_max@inbox.ru>
> Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
> Cc: Yann E. MORIN <yann.morin.1998@free.fr>
> ---
> This is probably not the definitive fix, but it's a step forward to
> understand the issue.
> 
> v2: fix typo in gcc/c/Make-lang.in
> add xgcc$(exeext) in C_SELFTEST_DEPS not in C_SELFTEST_FLAGS
> ---
>  ...gcc-c-cp-add-missing-xgcc-dependency.patch | 72 +++++++++++++++++++
>  1 file changed, 72 insertions(+)
>  create mode 100644 package/gcc/9.3.0/0004-gcc-c-cp-add-missing-xgcc-dependency.patch
> 
> diff --git a/package/gcc/9.3.0/0004-gcc-c-cp-add-missing-xgcc-dependency.patch b/package/gcc/9.3.0/0004-gcc-c-cp-add-missing-xgcc-dependency.patch
> new file mode 100644
> index 0000000000..172cff9ab9
> --- /dev/null
> +++ b/package/gcc/9.3.0/0004-gcc-c-cp-add-missing-xgcc-dependency.patch
> @@ -0,0 +1,72 @@
> +From 06aa7c5b5495aa1470e6cda5ebcae0a8bdbef182 Mon Sep 17 00:00:00 2001
> +From: Romain Naour <romain.naour@gmail.com>
> +Date: Wed, 20 May 2020 22:55:38 +0200
> +Subject: [PATCH] gcc/{c,cp}: add missing xgcc dependency
> +
> +As remported by several Buildroot users [1][2][3], the gcc build
> +fail while running selftests makefile target.
> +
> +The problem only occurs when ccache is used with gcc 9 and 10,
> +probably due to a race condition.
> +
> +While debuging with "make -p" we can notice that s-selftest-c target
> +contain only "cc1" as dependency instead of cc1 and SELFTEST_DEPS [4].
> +
> +  s-selftest-c: cc1

The thing is (using gcc master as reference):

  - SELFTEST_DEPS is used in gcc/c/Make-lang.in to generate a dependency
    rule:
        120 C_SELFTEST_DEPS = cc1$(exeext) $(SELFTEST_DEPS)
        123 s-selftest-c: $(C_SELFTEST_DEPS)

  - gcc/c/Make-lang.in is included from gcc/Makefile.in on line 1757:
        1757 include $(LANG_MAKEFRAGS)

  - SELFTESTS_DEPS is defined in gcc/Makefile.in line on 2009:
        2009 SELFTEST_DEPS = $(GCC_PASSES) stmp-int-hdrs $(srcdir)/testsuite/selftests

So, at the time the dependency rule on line 123 in gcc/c/Make-lang.in is
generated, SELFTEST_DEPS is not yet set, hence the dependency is not
generated.

So I think the proper solution would be to have SELFTEST_DEPS, as well
as all the variables in uses, defined before being used, i.e. before line
1757 in gcc/Makefile.

Notes:

  - LANG_MAKEFRAGS is set in gcc/Makefile.in:
        1129 LANG_MAKEFRAGS = @all_lang_makefrags@

  - all_lang_makefrags is substitued from gcc/configure.ac, and set to
    list fragments in all enabled sub-directories:
        6499     all_lang_makefrags="$all_lang_makefrags \$(srcdir)/$gcc_subdir/Make-lang.in"

And commit 033eb56717 is indeed the culprit: before it, the per-language
selftests were defined after SELFTEST_DEPS is defined, but moving these
test to the per-language fragments actually moved them before the
definition of SELFTEST_DEPS.

Regards,
Yann E. MORIN.

> +While the build is failing, the s-selftest-c dependencies recipe is
> +still running and reported as a bug by make.
> +
> +  "Dependencies recipe running (THIS IS A BUG)."
> +
> +A change [5] in gcc 9 seems to introduce the problem since we can't
> +reproduce this problem with gcc 8.
> +
> +For now add explicitely xgcc dependency to s-selftest-c and
> +s-selftest-c++. It not clear why SELFTEST_DEPS is lost.
> +
> +[1] http://lists.busybox.net/pipermail/buildroot/2020-May/282171.html
> +[2] http://lists.busybox.net/pipermail/buildroot/2020-May/282766.html
> +[3] https://github.com/cirosantilli/linux-kernel-module-cheat/issues/108
> +[4] https://gcc.gnu.org/git/?p=gcc.git;a=blob;f=gcc/c/Make-lang.in;h=bfae6fd2549c4f728816cd355fa9739dcc08fcde;hb=033eb5671769a4c681a44aad08a454e667e08502#l120
> +[5] https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=033eb5671769a4c681a44aad08a454e667e08502
> +
> +Signed-off-by: Romain Naour <romain.naour@gmail.com>
> +Cc: Ben Dakin-Norris <ben.dakin-norris@navtechradar.com>
> +Cc: Maxim Kochetkov <fido_max@inbox.ru>
> +Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
> +Cc: Yann E. MORIN <yann.morin.1998@free.fr>
> +---
> + gcc/c/Make-lang.in  | 2 +-
> + gcc/cp/Make-lang.in | 2 +-
> + 2 files changed, 2 insertions(+), 2 deletions(-)
> +
> +diff --git a/gcc/c/Make-lang.in b/gcc/c/Make-lang.in
> +index 1422be6e013..42f8a538f1a 100644
> +--- a/gcc/c/Make-lang.in
> ++++ b/gcc/c/Make-lang.in
> +@@ -117,7 +117,7 @@ c.srcman:
> + selftest-c: s-selftest-c
> + 
> + C_SELFTEST_FLAGS = -xc $(SELFTEST_FLAGS)
> +-C_SELFTEST_DEPS = cc1$(exeext) $(SELFTEST_DEPS)
> ++C_SELFTEST_DEPS = cc1$(exeext) xgcc$(exeext) $(SELFTEST_DEPS)
> + 
> + # Run the C selftests:
> + s-selftest-c: $(C_SELFTEST_DEPS)
> +diff --git a/gcc/cp/Make-lang.in b/gcc/cp/Make-lang.in
> +index 8fc1570659e..b2412544756 100644
> +--- a/gcc/cp/Make-lang.in
> ++++ b/gcc/cp/Make-lang.in
> +@@ -170,7 +170,7 @@ c++.srcman: doc/g++.1
> + # at each stage of the build:
> + selftest-c++: s-selftest-c++
> + 
> +-CPP_SELFTEST_DEPS = cc1plus$(exeext) $(SELFTEST_DEPS)
> ++CPP_SELFTEST_DEPS = cc1plus$(exeext) xgcc$(exeext) $(SELFTEST_DEPS)
> + CPP_SELFTEST_FLAGS = -xc++ $(SELFTEST_FLAGS)
> + 
> + # Run the C++ selftests
> +-- 
> +2.25.4
> +
> -- 
> 2.25.4
> 

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

  reply	other threads:[~2020-05-21 12:12 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-20 22:45 [Buildroot] [PATCHv2] package/gcc/9.3.0: fix host-gcc-final when ccache is used Romain Naour
2020-05-21 12:12 ` Yann E. MORIN [this message]
2020-05-21 12:38   ` Yann E. MORIN

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200521121228.GM27030@scaer \
    --to=yann.morin.1998@free.fr \
    --cc=buildroot@busybox.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.