* [PATCH 12/14] [UML] fix gcc ICEs and unresolved externs
@ 2008-08-01 22:41 Jeff Dike
0 siblings, 0 replies; 3+ messages in thread
From: Jeff Dike @ 2008-08-01 22:41 UTC (permalink / raw)
To: stable
Cc: LKML, uml-devel, Benny Halevy, Adrian Bunk, Ingo Molnar,
Ulrich Drepper
>From f9498bcc86dbf5249a4d9baff6ba508c3d4c86fa Mon Sep 17 00:00:00 2001
There are various constraints on the use of unit-at-a-time:
- i386 uses no-unit-at-a-time for pre-4.0 (not 4.3)
- x86_64 uses unit-at-a-time always
Uli reported a crash on x86_64 with gcc 4.1.2 with unit-at-a-time,
resulting in commit c0a18111e571138747a98af18b3a2124df56a0d1
Ingo reported a gcc internal error with gcc 4.3 with no-unit-at-a-timem,
resulting in 22eecde2f9034764a3fd095eecfa3adfb8ec9a98
Benny Halevy is seeing extern inlines not resolved with gcc 4.3 with
no-unit-at-a-time
This patch reintroduces unit-at-a-time for gcc >= 4.0, bringing back the
possibility of Uli's crash. If that happens, we'll debug it.
I started seeing both the internal compiler errors and unresolved
inlines on Fedora 9. This patch fixes both problems, without so far
reintroducing the crash reported by Uli.
Signed-off-by: Jeff Dike <jdike@linux.intel.com>
Cc: Benny Halevy <bhalevy@panasas.com>
Cc: Adrian Bunk <bunk@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
arch/um/Makefile | 1 -
arch/um/Makefile-i386 | 7 +++++++
arch/um/Makefile-x86_64 | 3 +++
3 files changed, 10 insertions(+), 1 deletions(-)
diff --git a/arch/um/Makefile b/arch/um/Makefile
index dbeab15..ca40397 100644
--- a/arch/um/Makefile
+++ b/arch/um/Makefile
@@ -77,7 +77,6 @@ include $(srctree)/$(ARCH_DIR)/Makefile-os-$(OS)
KERNEL_DEFINES = $(strip -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask \
-Dmktime=kernel_mktime $(ARCH_KERNEL_DEFINES))
KBUILD_CFLAGS += $(KERNEL_DEFINES)
-KBUILD_CFLAGS += $(call cc-option,-fno-unit-at-a-time,)
PHONY += linux
diff --git a/arch/um/Makefile-i386 b/arch/um/Makefile-i386
index 561e373..302cbe5 100644
--- a/arch/um/Makefile-i386
+++ b/arch/um/Makefile-i386
@@ -32,4 +32,11 @@ cflags-y += $(call cc-option,-mpreferred-stack-boundary=2)
# an unresolved reference.
cflags-y += -ffreestanding
+# Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use
+# a lot more stack due to the lack of sharing of stacklots. Also, gcc
+# 4.3.0 needs -funit-at-a-time for extern inline functions.
+KBUILD_CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then \
+ echo $(call cc-option,-fno-unit-at-a-time); \
+ else echo $(call cc-option,-funit-at-a-time); fi ;)
+
KBUILD_CFLAGS += $(cflags-y)
diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64
index 8ed362f..a9cd7e7 100644
--- a/arch/um/Makefile-x86_64
+++ b/arch/um/Makefile-x86_64
@@ -21,3 +21,6 @@ HEADER_ARCH := x86
LINK-$(CONFIG_LD_SCRIPT_DYN) += -Wl,-rpath,/lib64
LINK-y += -m64
+
+# Do unit-at-a-time unconditionally on x86_64, following the host
+KBUILD_CFLAGS += $(call cc-option,-funit-at-a-time)
--
1.5.5.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* [PATCH 12/14] [UML] fix gcc ICEs and unresolved externs
@ 2008-08-05 20:14 Jeff Dike
2008-08-16 22:32 ` patch uml-fix-gcc-ices-and-unresolved-externs.patch added to 2.6.25-stable tree gregkh
0 siblings, 1 reply; 3+ messages in thread
From: Jeff Dike @ 2008-08-05 20:14 UTC (permalink / raw)
To: stable; +Cc: LKML, uml-devel
>From 4f81c5350b44bcc501ab6f8a089b16d064b4d2f6
There are various constraints on the use of unit-at-a-time:
- i386 uses no-unit-at-a-time for pre-4.0 (not 4.3)
- x86_64 uses unit-at-a-time always
Uli reported a crash on x86_64 with gcc 4.1.2 with unit-at-a-time,
resulting in commit c0a18111e571138747a98af18b3a2124df56a0d1
Ingo reported a gcc internal error with gcc 4.3 with no-unit-at-a-timem,
resulting in 22eecde2f9034764a3fd095eecfa3adfb8ec9a98
Benny Halevy is seeing extern inlines not resolved with gcc 4.3 with
no-unit-at-a-time
This patch reintroduces unit-at-a-time for gcc >= 4.0, bringing back the
possibility of Uli's crash. If that happens, we'll debug it.
I started seeing both the internal compiler errors and unresolved
inlines on Fedora 9. This patch fixes both problems, without so far
reintroducing the crash reported by Uli.
Signed-off-by: Jeff Dike <jdike@linux.intel.com>
Cc: Benny Halevy <bhalevy@panasas.com>
Cc: Adrian Bunk <bunk@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
arch/um/Makefile | 1 -
arch/um/Makefile-i386 | 7 +++++++
arch/um/Makefile-x86_64 | 3 +++
3 files changed, 10 insertions(+), 1 deletions(-)
diff --git a/arch/um/Makefile b/arch/um/Makefile
index dbeab15..ca40397 100644
--- a/arch/um/Makefile
+++ b/arch/um/Makefile
@@ -77,7 +77,6 @@ include $(srctree)/$(ARCH_DIR)/Makefile-os-$(OS)
KERNEL_DEFINES = $(strip -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask \
-Dmktime=kernel_mktime $(ARCH_KERNEL_DEFINES))
KBUILD_CFLAGS += $(KERNEL_DEFINES)
-KBUILD_CFLAGS += $(call cc-option,-fno-unit-at-a-time,)
PHONY += linux
diff --git a/arch/um/Makefile-i386 b/arch/um/Makefile-i386
index 561e373..302cbe5 100644
--- a/arch/um/Makefile-i386
+++ b/arch/um/Makefile-i386
@@ -32,4 +32,11 @@ cflags-y += $(call cc-option,-mpreferred-stack-boundary=2)
# an unresolved reference.
cflags-y += -ffreestanding
+# Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use
+# a lot more stack due to the lack of sharing of stacklots. Also, gcc
+# 4.3.0 needs -funit-at-a-time for extern inline functions.
+KBUILD_CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then \
+ echo $(call cc-option,-fno-unit-at-a-time); \
+ else echo $(call cc-option,-funit-at-a-time); fi ;)
+
KBUILD_CFLAGS += $(cflags-y)
diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64
index 8ed362f..a9cd7e7 100644
--- a/arch/um/Makefile-x86_64
+++ b/arch/um/Makefile-x86_64
@@ -21,3 +21,6 @@ HEADER_ARCH := x86
LINK-$(CONFIG_LD_SCRIPT_DYN) += -Wl,-rpath,/lib64
LINK-y += -m64
+
+# Do unit-at-a-time unconditionally on x86_64, following the host
+KBUILD_CFLAGS += $(call cc-option,-funit-at-a-time)
--
1.5.5.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* patch uml-fix-gcc-ices-and-unresolved-externs.patch added to 2.6.25-stable tree
2008-08-05 20:14 [PATCH 12/14] [UML] fix gcc ICEs and unresolved externs Jeff Dike
@ 2008-08-16 22:32 ` gregkh
0 siblings, 0 replies; 3+ messages in thread
From: gregkh @ 2008-08-16 22:32 UTC (permalink / raw)
To: jdike, bhalevy, bunk, drepper, gregkh, jdike, linux-kernel, mingo,
torvalds, user-mode-linux-devel
Cc: stable, stable-commits
This is a note to let you know that we have just queued up the patch titled
Subject: uml: fix gcc ICEs and unresolved externs
to the 2.6.25-stable tree. Its filename is
uml-fix-gcc-ices-and-unresolved-externs.patch
A git repo of this tree can be found at
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
>From stable-bounces@linux.kernel.org Tue Aug 5 13:14:46 2008
From: Jeff Dike <jdike@addtoit.com>
Date: Tue, 5 Aug 2008 16:14:10 -0400
Subject: uml: fix gcc ICEs and unresolved externs
To: stable@kernel.org
Cc: LKML <linux-kernel@vger.kernel.org>, uml-devel <user-mode-linux-devel@lists.sourceforge.net>
Message-ID: <20080805201410.GA11794@c2.user-mode-linux.org>
Content-Disposition: inline
From: Jeff Dike <jdike@addtoit.com>
commit 4f81c5350b44bcc501ab6f8a089b16d064b4d2f6 upstream
There are various constraints on the use of unit-at-a-time:
- i386 uses no-unit-at-a-time for pre-4.0 (not 4.3)
- x86_64 uses unit-at-a-time always
Uli reported a crash on x86_64 with gcc 4.1.2 with unit-at-a-time,
resulting in commit c0a18111e571138747a98af18b3a2124df56a0d1
Ingo reported a gcc internal error with gcc 4.3 with no-unit-at-a-timem,
resulting in 22eecde2f9034764a3fd095eecfa3adfb8ec9a98
Benny Halevy is seeing extern inlines not resolved with gcc 4.3 with
no-unit-at-a-time
This patch reintroduces unit-at-a-time for gcc >= 4.0, bringing back the
possibility of Uli's crash. If that happens, we'll debug it.
I started seeing both the internal compiler errors and unresolved
inlines on Fedora 9. This patch fixes both problems, without so far
reintroducing the crash reported by Uli.
Signed-off-by: Jeff Dike <jdike@linux.intel.com>
Cc: Benny Halevy <bhalevy@panasas.com>
Cc: Adrian Bunk <bunk@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
arch/um/Makefile | 1 -
arch/um/Makefile-i386 | 7 +++++++
arch/um/Makefile-x86_64 | 3 +++
3 files changed, 10 insertions(+), 1 deletion(-)
--- a/arch/um/Makefile
+++ b/arch/um/Makefile
@@ -77,7 +77,6 @@ include $(srctree)/$(ARCH_DIR)/Makefile-
KERNEL_DEFINES = $(strip -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask \
-Dmktime=kernel_mktime $(ARCH_KERNEL_DEFINES))
KBUILD_CFLAGS += $(KERNEL_DEFINES)
-KBUILD_CFLAGS += $(call cc-option,-fno-unit-at-a-time,)
PHONY += linux
--- a/arch/um/Makefile-i386
+++ b/arch/um/Makefile-i386
@@ -32,4 +32,11 @@ cflags-y += $(call cc-option,-mpreferred
# an unresolved reference.
cflags-y += -ffreestanding
+# Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use
+# a lot more stack due to the lack of sharing of stacklots. Also, gcc
+# 4.3.0 needs -funit-at-a-time for extern inline functions.
+KBUILD_CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then \
+ echo $(call cc-option,-fno-unit-at-a-time); \
+ else echo $(call cc-option,-funit-at-a-time); fi ;)
+
KBUILD_CFLAGS += $(cflags-y)
--- a/arch/um/Makefile-x86_64
+++ b/arch/um/Makefile-x86_64
@@ -21,3 +21,6 @@ HEADER_ARCH := x86
LINK-$(CONFIG_LD_SCRIPT_DYN) += -Wl,-rpath,/lib64
LINK-y += -m64
+
+# Do unit-at-a-time unconditionally on x86_64, following the host
+KBUILD_CFLAGS += $(call cc-option,-funit-at-a-time)
Patches currently in stable-queue which might be from jdike@addtoit.com are
queue-2.6.25/uml-fix-build-when-slob-is-enabled.patch
queue-2.6.25/uml-fix-bad-ntp-interaction-with-clock.patch
queue-2.6.25/uml-physical-memory-shouldn-t-include-initial-stack.patch
queue-2.6.25/uml-track-and-make-up-lost-ticks.patch
queue-2.6.25/uml-missed-kmalloc-in-pcap_user.c.patch
queue-2.6.25/uml-deal-with-host-time-going-backwards.patch
queue-2.6.25/uml-deal-with-inaccessible-address-space-start.patch
queue-2.6.25/uml-missing-export-of-csum_partial-on-uml-amd64.patch
queue-2.6.25/uml-memcpy-export-needs-to-follow-host-declaration.patch
queue-2.6.25/uml-stub-needs-to-tolerate-sigwinch.patch
queue-2.6.25/uml-work-around-broken-host-ptrace_sysemu.patch
queue-2.6.25/uml-fix-gcc-ices-and-unresolved-externs.patch
queue-2.6.25/uml-fix-boot-crash.patch
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2008-08-16 22:37 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-08-05 20:14 [PATCH 12/14] [UML] fix gcc ICEs and unresolved externs Jeff Dike
2008-08-16 22:32 ` patch uml-fix-gcc-ices-and-unresolved-externs.patch added to 2.6.25-stable tree gregkh
-- strict thread matches above, loose matches on Subject: below --
2008-08-01 22:41 [PATCH 12/14] [UML] fix gcc ICEs and unresolved externs Jeff Dike
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox