All of lore.kernel.org
 help / color / mirror / Atom feed
From: Arnd Bergmann <arnd@arndb.de>
To: linux-arch@vger.kernel.org
Cc: linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@vger.kernel.org,
	Russell King <rmk+kernel@armlinux.org.uk>,
	Andrew Morton <akpm@linux-foundation.org>,
	kernel-build-reports@lists.linaro.org,
	kvmarm@lists.cs.columbia.edu, Arnd Bergmann <arnd@arndb.de>
Subject: [PATCH 03/13] ARM: div64: fix building with gcc-4.5 and lower
Date: Fri, 16 Dec 2016 11:56:24 +0100	[thread overview]
Message-ID: <20161216105634.235457-4-arnd@arndb.de> (raw)
In-Reply-To: <20161216105634.235457-1-arnd@arndb.de>

drivers/media/dvb-frontends/s921.c: In function 's921_pll_tune':
arch/arm/include/asm/div64.h:109: error: can't find a register in class 'GENERAL_REGS' while reloading 'asm'
arch/arm/include/asm/div64.h:127: error: can't find a register in class 'GENERAL_REGS' while reloading 'asm'
arch/arm/include/asm/div64.h:109: error: 'asm' operand has impossible constraints
arch/arm/include/asm/div64.h:127: error: 'asm' operand has impossible constraints
scripts/Makefile.build:299: recipe for target 'drivers/media/dvb-frontends/s921.o' failed

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/arm/include/asm/div64.h | 17 +++--------------
 1 file changed, 3 insertions(+), 14 deletions(-)

diff --git a/arch/arm/include/asm/div64.h b/arch/arm/include/asm/div64.h
index 7ca0c613c33b..e1c8d4a58797 100644
--- a/arch/arm/include/asm/div64.h
+++ b/arch/arm/include/asm/div64.h
@@ -66,27 +66,16 @@ static inline uint32_t __div64_32(uint64_t *n, uint32_t base)
  */
 #define do_div(n, base) __div64_32(&(n), base)
 
-#elif defined(CONFIG_CPU_32v3)
+#elif GCC_VERSION < 40600
 
 /*
- * modern compiler versions (>= gcc-4.9) tend to misoptimize
- * the code for ARMv3, and this is not getting fixed any more.
+ * gcc-4.4 and gcc-4.5 tend to run out of registers in their
+ * register allocation in some functions
  */
 #define do_div(n, base) __div64_32(&(n), base)
 
 #else
 
-/*
- * gcc versions earlier than 4.0 are simply too problematic for the
- * __div64_const32() code in asm-generic/div64.h. First there is
- * gcc PR 15089 that tend to trig on more complex constructs, spurious
- * .global __udivsi3 are inserted even if none of those symbols are
- * referenced in the generated code, and those gcc versions are not able
- * to do constant propagation on long long values anyway.
- */
-
-#define __div64_const32_is_OK (__GNUC__ >= 4)
-
 static inline uint64_t __arch_xprod_64(uint64_t m, uint64_t n, bool bias)
 {
 	unsigned long long res;
-- 
2.9.0

WARNING: multiple messages have this Message-ID (diff)
From: Arnd Bergmann <arnd@arndb.de>
To: linux-arch@vger.kernel.org
Cc: Arnd Bergmann <arnd@arndb.de>,
	kernel-build-reports@lists.linaro.org,
	linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org,
	Russell King <rmk+kernel@armlinux.org.uk>,
	Andrew Morton <akpm@linux-foundation.org>,
	kvmarm@lists.cs.columbia.edu, linux-arm-kernel@vger.kernel.org
Subject: [PATCH 03/13] ARM: div64: fix building with gcc-4.5 and lower
Date: Fri, 16 Dec 2016 11:56:24 +0100	[thread overview]
Message-ID: <20161216105634.235457-4-arnd@arndb.de> (raw)
In-Reply-To: <20161216105634.235457-1-arnd@arndb.de>

drivers/media/dvb-frontends/s921.c: In function 's921_pll_tune':
arch/arm/include/asm/div64.h:109: error: can't find a register in class 'GENERAL_REGS' while reloading 'asm'
arch/arm/include/asm/div64.h:127: error: can't find a register in class 'GENERAL_REGS' while reloading 'asm'
arch/arm/include/asm/div64.h:109: error: 'asm' operand has impossible constraints
arch/arm/include/asm/div64.h:127: error: 'asm' operand has impossible constraints
scripts/Makefile.build:299: recipe for target 'drivers/media/dvb-frontends/s921.o' failed

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/arm/include/asm/div64.h | 17 +++--------------
 1 file changed, 3 insertions(+), 14 deletions(-)

diff --git a/arch/arm/include/asm/div64.h b/arch/arm/include/asm/div64.h
index 7ca0c613c33b..e1c8d4a58797 100644
--- a/arch/arm/include/asm/div64.h
+++ b/arch/arm/include/asm/div64.h
@@ -66,27 +66,16 @@ static inline uint32_t __div64_32(uint64_t *n, uint32_t base)
  */
 #define do_div(n, base) __div64_32(&(n), base)
 
-#elif defined(CONFIG_CPU_32v3)
+#elif GCC_VERSION < 40600
 
 /*
- * modern compiler versions (>= gcc-4.9) tend to misoptimize
- * the code for ARMv3, and this is not getting fixed any more.
+ * gcc-4.4 and gcc-4.5 tend to run out of registers in their
+ * register allocation in some functions
  */
 #define do_div(n, base) __div64_32(&(n), base)
 
 #else
 
-/*
- * gcc versions earlier than 4.0 are simply too problematic for the
- * __div64_const32() code in asm-generic/div64.h. First there is
- * gcc PR 15089 that tend to trig on more complex constructs, spurious
- * .global __udivsi3 are inserted even if none of those symbols are
- * referenced in the generated code, and those gcc versions are not able
- * to do constant propagation on long long values anyway.
- */
-
-#define __div64_const32_is_OK (__GNUC__ >= 4)
-
 static inline uint64_t __arch_xprod_64(uint64_t m, uint64_t n, bool bias)
 {
 	unsigned long long res;
-- 
2.9.0

  parent reply	other threads:[~2016-12-16 10:59 UTC|newest]

Thread overview: 113+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-16 10:56 [RFC] minimum gcc version for kernel: raise to gcc-4.3 or 4.6? Arnd Bergmann
2016-12-16 10:56 ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 01/13] [HACK] gcc-4.5: avoid link errors for unused function pointers Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 02/13] KVM: arm: fix gcc-4.5 build Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2017-01-04 10:38   ` Christoffer Dall
2017-01-04 10:38     ` Christoffer Dall
2016-12-16 10:56 ` Arnd Bergmann [this message]
2016-12-16 10:56   ` [PATCH 03/13] ARM: div64: fix building with gcc-4.5 and lower Arnd Bergmann
2016-12-16 10:56 ` [PATCH 04/13] vfio-pci: use 32-bit comparisons for register address for gcc-4.5 Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 15:30   ` Alex Williamson
2016-12-16 19:50     ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 05/13] clk: pxa: fix gcc-4.4 build Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 06/13] ARM: atomic: " Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 07/13] watchdog: kempld: fix gcc-4.3 build Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 08/13] arm/arm64: xen: avoid gcc-4.4 warning Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 09/13] ARM: mark cmpxchg and xchg __always_inline for gcc-4.3 Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 10/13] asm-generic: mark cmpxchg as " Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 10:56 ` [PATCH 11/13] fs: fix unsigned enum warning with gcc-4.2 Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2017-01-03 22:47   ` Brendan Gregg
2017-02-28 21:53     ` Brendan Gregg
2016-12-16 10:56 ` [PATCH 12/13] KVM: arm: avoid binary number literals for gcc-4.2 Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2017-01-04 10:39   ` Christoffer Dall
2017-01-04 10:39     ` Christoffer Dall
2016-12-16 10:56 ` [PATCH 13/13] ARM: avoid 'Q' asm constraint for gcc-4.1 and earlier Arnd Bergmann
2016-12-16 10:56   ` Arnd Bergmann
2016-12-16 11:14 ` [RFC] minimum gcc version for kernel: raise to gcc-4.3 or 4.6? Arnd Bergmann
2016-12-16 11:14   ` Arnd Bergmann
2016-12-16 11:14   ` Arnd Bergmann
2017-04-16 19:52   ` Kees Cook
2017-04-16 19:52     ` Kees Cook
2017-04-16 19:52     ` Kees Cook
2017-04-20 10:15     ` Arnd Bergmann
2017-04-20 10:15       ` Arnd Bergmann
2017-04-20 10:15       ` Arnd Bergmann
2017-04-20 19:52       ` Kees Cook
2017-04-20 19:52         ` Kees Cook
2017-04-20 19:52         ` Kees Cook
2017-04-20 19:52         ` Kees Cook
2017-04-21 20:55         ` Arnd Bergmann
2017-04-21 20:55           ` Arnd Bergmann
2017-04-21 20:55           ` Arnd Bergmann
2017-04-21 21:05           ` Kees Cook
2017-04-21 21:05             ` Kees Cook
2017-04-21 21:05             ` Kees Cook
2017-04-22  3:10             ` Maciej W. Rozycki
2017-04-22  3:10               ` Maciej W. Rozycki
2017-04-22  3:10               ` Maciej W. Rozycki
2017-04-22 15:30               ` Arnd Bergmann
2017-04-22 15:30                 ` Arnd Bergmann
2017-04-22 15:30                 ` Arnd Bergmann
2017-04-23 20:13                 ` Geert Uytterhoeven
2017-04-23 20:13                   ` Geert Uytterhoeven
2017-04-23 20:13                   ` Geert Uytterhoeven
2017-04-23 20:13                   ` Geert Uytterhoeven
2017-04-24  9:44                   ` Arnd Bergmann
2017-04-24  9:44                     ` Arnd Bergmann
2017-04-24  9:44                     ` Arnd Bergmann
2017-04-24  9:44                     ` Arnd Bergmann
2017-04-24 10:17                     ` Geert Uytterhoeven
2017-04-24 10:17                       ` Geert Uytterhoeven
2017-04-24 10:17                       ` Geert Uytterhoeven
2017-04-24 10:17                       ` Geert Uytterhoeven
2017-04-24 14:13                       ` Arnd Bergmann
2017-04-24 14:13                         ` Arnd Bergmann
2017-04-24 14:13                         ` Arnd Bergmann
2017-04-24 16:53                     ` Maciej W. Rozycki
2017-04-24 16:53                       ` Maciej W. Rozycki
2017-04-24 16:53                       ` Maciej W. Rozycki
2017-04-24 17:29                       ` Arnd Bergmann
2017-04-24 17:29                         ` Arnd Bergmann
2017-04-24 17:29                         ` Arnd Bergmann
2017-04-24 18:16                         ` Geert Uytterhoeven
2017-04-24 18:16                           ` Geert Uytterhoeven
2017-04-24 18:16                           ` Geert Uytterhoeven
2017-04-24 18:16                           ` Geert Uytterhoeven
2017-04-24 18:30                         ` Maciej W. Rozycki
2017-04-24 18:30                           ` Maciej W. Rozycki
2017-04-24 18:30                           ` Maciej W. Rozycki
2017-04-24 18:30                           ` Maciej W. Rozycki
2017-04-24 20:30                           ` Arnd Bergmann
2017-04-24 20:30                             ` Arnd Bergmann
2017-04-24 20:30                             ` Arnd Bergmann
2017-04-24 20:52                             ` Kees Cook
2017-04-24 20:52                               ` Kees Cook
2017-04-24 20:52                               ` Kees Cook
2017-04-25  7:06                               ` Geert Uytterhoeven
2017-04-25  7:06                                 ` Geert Uytterhoeven
2017-04-25  7:06                                 ` Geert Uytterhoeven
2017-04-25  9:22     ` Suzuki K Poulose
2017-04-25  9:22       ` Suzuki K Poulose
2017-04-25  9:22       ` Suzuki K Poulose
2016-12-16 15:54 ` Geert Uytterhoeven
2016-12-16 15:54   ` Geert Uytterhoeven
2016-12-16 19:58   ` Arnd Bergmann
2016-12-16 19:58     ` Arnd Bergmann
2016-12-16 20:34     ` Geert Uytterhoeven
2016-12-16 17:00 ` Sebastian Andrzej Siewior
2016-12-16 22:00   ` Arnd Bergmann
2016-12-16 22:00     ` Arnd Bergmann
2016-12-17 11:29     ` Sebastian Andrzej Siewior
2017-01-02 12:23       ` Russell King - ARM Linux
2016-12-20  9:59     ` Heiko Carstens

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=20161216105634.235457-4-arnd@arndb.de \
    --to=arnd@arndb.de \
    --cc=akpm@linux-foundation.org \
    --cc=kernel-build-reports@lists.linaro.org \
    --cc=kvmarm@lists.cs.columbia.edu \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-arm-kernel@vger.kernel.org \
    --cc=linux-kbuild@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rmk+kernel@armlinux.org.uk \
    /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.