All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marc Zyngier <maz@kernel.org>
To: linux-arm-kernel@lists.infradead.org, kvmarm@lists.cs.columbia.edu
Cc: Catalin Marinas <catalin.marinas@arm.com>,
	Will Deacon <will@kernel.org>, Ard Biesheuvel <ardb@kernel.org>,
	kernel-team@android.com
Subject: [PATCH] arm64: Work around broken GCC 4.9 handling of "S" constraint
Date: Thu, 17 Dec 2020 11:11:35 +0000	[thread overview]
Message-ID: <20201217111135.1536658-1-maz@kernel.org> (raw)

GCC 4.9 seems to have a problem with the "S" asm constraint
when the symbol lives in the same compilation unit, and pretends
the constraint is impossible:

$ cat x.c
void *foo(void)
{
	static int x;
	int *addr;
	asm("adrp %0, %1" : "=r" (addr) : "S" (&x));
	return addr;
}

$ ~/Work/gcc-linaro-aarch64-linux-gnu-4.9-2014.09_linux/bin/aarch64-linux-gnu-gcc -S -x c -O2 x.c
x.c: In function ‘foo’:
x.c:5:2: error: impossible constraint in ‘asm’
  asm("adrp %0, %1" : "=r" (addr) : "S" (&x));
  ^

Boo. Following revisions of the compiler work just fine, though.

We can fallback to the "i" constraint for GCC version prior to 5.0,
which *seems* to do the right thing. Hopefully we will be able to
remove this at some point, but in the meantime this gets us going.

Signed-off-by: Marc Zyngier <maz@kernel.org>
---
* From v1: Dropped the detection hack and rely on GCC_VERSION

 arch/arm64/include/asm/kvm_asm.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h
index 7ccf770c53d9..8a33d83ea843 100644
--- a/arch/arm64/include/asm/kvm_asm.h
+++ b/arch/arm64/include/asm/kvm_asm.h
@@ -199,6 +199,12 @@ extern void __vgic_v3_init_lrs(void);
 
 extern u32 __kvm_get_mdcr_el2(void);
 
+#if defined(GCC_VERSION) && GCC_VERSION < 50000
+#define SYM_CONSTRAINT	"i"
+#else
+#define SYM_CONSTRAINT	"S"
+#endif
+
 /*
  * Obtain the PC-relative address of a kernel symbol
  * s: symbol
@@ -215,7 +221,7 @@ extern u32 __kvm_get_mdcr_el2(void);
 		typeof(s) *addr;					\
 		asm("adrp	%0, %1\n"				\
 		    "add	%0, %0, :lo12:%1\n"			\
-		    : "=r" (addr) : "S" (&s));				\
+		    : "=r" (addr) : SYM_CONSTRAINT (&s));		\
 		addr;							\
 	})
 
-- 
2.29.2

_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm

WARNING: multiple messages have this Message-ID (diff)
From: Marc Zyngier <maz@kernel.org>
To: linux-arm-kernel@lists.infradead.org, kvmarm@lists.cs.columbia.edu
Cc: Catalin Marinas <catalin.marinas@arm.com>,
	David Brazdil <dbrazdil@google.com>,
	Will Deacon <will@kernel.org>, Ard Biesheuvel <ardb@kernel.org>,
	kernel-team@android.com
Subject: [PATCH] arm64: Work around broken GCC 4.9 handling of "S" constraint
Date: Thu, 17 Dec 2020 11:11:35 +0000	[thread overview]
Message-ID: <20201217111135.1536658-1-maz@kernel.org> (raw)

GCC 4.9 seems to have a problem with the "S" asm constraint
when the symbol lives in the same compilation unit, and pretends
the constraint is impossible:

$ cat x.c
void *foo(void)
{
	static int x;
	int *addr;
	asm("adrp %0, %1" : "=r" (addr) : "S" (&x));
	return addr;
}

$ ~/Work/gcc-linaro-aarch64-linux-gnu-4.9-2014.09_linux/bin/aarch64-linux-gnu-gcc -S -x c -O2 x.c
x.c: In function ‘foo’:
x.c:5:2: error: impossible constraint in ‘asm’
  asm("adrp %0, %1" : "=r" (addr) : "S" (&x));
  ^

Boo. Following revisions of the compiler work just fine, though.

We can fallback to the "i" constraint for GCC version prior to 5.0,
which *seems* to do the right thing. Hopefully we will be able to
remove this at some point, but in the meantime this gets us going.

Signed-off-by: Marc Zyngier <maz@kernel.org>
---
* From v1: Dropped the detection hack and rely on GCC_VERSION

 arch/arm64/include/asm/kvm_asm.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h
index 7ccf770c53d9..8a33d83ea843 100644
--- a/arch/arm64/include/asm/kvm_asm.h
+++ b/arch/arm64/include/asm/kvm_asm.h
@@ -199,6 +199,12 @@ extern void __vgic_v3_init_lrs(void);
 
 extern u32 __kvm_get_mdcr_el2(void);
 
+#if defined(GCC_VERSION) && GCC_VERSION < 50000
+#define SYM_CONSTRAINT	"i"
+#else
+#define SYM_CONSTRAINT	"S"
+#endif
+
 /*
  * Obtain the PC-relative address of a kernel symbol
  * s: symbol
@@ -215,7 +221,7 @@ extern u32 __kvm_get_mdcr_el2(void);
 		typeof(s) *addr;					\
 		asm("adrp	%0, %1\n"				\
 		    "add	%0, %0, :lo12:%1\n"			\
-		    : "=r" (addr) : "S" (&s));				\
+		    : "=r" (addr) : SYM_CONSTRAINT (&s));		\
 		addr;							\
 	})
 
-- 
2.29.2


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

             reply	other threads:[~2020-12-17 11:11 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-17 11:11 Marc Zyngier [this message]
2020-12-17 11:11 ` [PATCH] arm64: Work around broken GCC 4.9 handling of "S" constraint Marc Zyngier
2020-12-17 11:13 ` Ard Biesheuvel
2020-12-17 11:13   ` Ard Biesheuvel
2020-12-17 14:13 ` Catalin Marinas
2020-12-17 14:13   ` Catalin Marinas

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=20201217111135.1536658-1-maz@kernel.org \
    --to=maz@kernel.org \
    --cc=ardb@kernel.org \
    --cc=catalin.marinas@arm.com \
    --cc=kernel-team@android.com \
    --cc=kvmarm@lists.cs.columbia.edu \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=will@kernel.org \
    /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.