diff mbox

gcc trunk fails to build kernel on PowerPC64 due to oprofile warnings

Message ID fe2c2fd9-d255-61ca-5328-ca5d45827f47@redhat.com
State New
Headers show

Commit Message

William Cohen Jan. 26, 2017, 3:46 p.m. UTC
On 01/25/2017 06:00 PM, Anton Blanchard wrote:
> Hi,
> 
> gcc trunk has failed to build PowerPC64 kernels for a month or so. The issue
> is in oprofile, which is common code but ends up being sucked into
> arch/powerpc and therefore subject to the -Werror applied to arch/powerpc:
>  
> linux/arch/powerpc/oprofile/../../../drivers/oprofile/oprofile_stats.c: In function ‘oprofile_create_stats_files’:
> linux/arch/powerpc/oprofile/../../../drivers/oprofile/oprofile_stats.c:55:25: error: ‘%d’ directive output may be truncated writing between 1 and 11 bytes into a region of size 7 [-Werror=format-truncation=]
>    snprintf(buf, 10, "cpu%d", i);
>                          ^~
> linux/arch/powerpc/oprofile/../../../drivers/oprofile/oprofile_stats.c:55:21: note: using the range [1, -2147483648] for directive argument
>    snprintf(buf, 10, "cpu%d", i);
>                      ^~~~~~~
> linux/arch/powerpc/oprofile/../../../drivers/oprofile/oprofile_stats.c:55:3: note: format output between 5 and 15 bytes into a destination of size 10
>    snprintf(buf, 10, "cpu%d", i);
>    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>   LD      crypto/async_tx/built-in.o
>   CC      lib/random32.o
> cc1: all warnings being treated as errors
> 
> Anton
> 
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> _______________________________________________
> oprofile-list mailing list
> oprofile-list@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/oprofile-list
> 

The attached patch should make the buffer large enough to eliminate the warning.

-Will

Comments

William Cohen Jan. 30, 2017, 3:35 p.m. UTC | #1
On 01/26/2017 11:06 AM, Robert Richter wrote:
> On 26.01.17 10:46:43, William Cohen wrote:
>> From 7e46dbd7dc5bc941926a4a63c28ccebf46493e8d Mon Sep 17 00:00:00 2001
>> From: William Cohen <wcohen@redhat.com>
>> Date: Thu, 26 Jan 2017 10:33:59 -0500
>> Subject: [PATCH] Avoid hypthetical string truncation in oprofile stats buffer
>> MIME-Version: 1.0
>> Content-Type: text/plain; charset=UTF-8
>> Content-Transfer-Encoding: 8bit
>>
>> Increased the size of an internal oprofile driver buffer ensuring that
>> the string was never truncated for any possible int value to avoid the
>> following gcc-7 compiler error on ppc when building the kernel:
> 
> Please test gcc7 for other archs first. I don't think this is the only
> change needed to avoid this warning in oprofile code.
> 
> Thanks,
> 
> -Robert

Hi Robert,

It took me a little while to get a gcc-7 environment set up with Fedora Rawhide on x86_64.  I have a kernel building now with gcc-7 and will see what other issues pop up during the compile.

While watching the output of the build I see the gcc-7 error message about truncating output of snprintf output seems to occur for a number of other places besides the oprofile kernel code during the compile.

-Will
diff mbox

Patch

From 7e46dbd7dc5bc941926a4a63c28ccebf46493e8d Mon Sep 17 00:00:00 2001
From: William Cohen <wcohen@redhat.com>
Date: Thu, 26 Jan 2017 10:33:59 -0500
Subject: [PATCH] Avoid hypthetical string truncation in oprofile stats buffer
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Increased the size of an internal oprofile driver buffer ensuring that
the string was never truncated for any possible int value to avoid the
following gcc-7 compiler error on ppc when building the kernel:

linux/arch/powerpc/oprofile/../../../drivers/oprofile/oprofile_stats.c: In function ‘oprofile_create_stats_files’:
linux/arch/powerpc/oprofile/../../../drivers/oprofile/oprofile_stats.c:55:25: error: ‘%d’ directive output may be truncated writing between 1 and 11 bytes into a region of size 7 [-Werror=format-truncation=]
   snprintf(buf, 10, "cpu%d", i);
                         ^~
linux/arch/powerpc/oprofile/../../../drivers/oprofile/oprofile_stats.c:55:21: note: using the range [1, -2147483648] for directive argument
   snprintf(buf, 10, "cpu%d", i);
                     ^~~~~~~
linux/arch/powerpc/oprofile/../../../drivers/oprofile/oprofile_stats.c:55:3: note: format output between 5 and 15 bytes into a destination of size 10
   snprintf(buf, 10, "cpu%d", i);
   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  LD      crypto/async_tx/built-in.o
  CC      lib/random32.o
cc1: all warnings being treated as errors

Signed-off-by: William Cohen <wcohen@redhat.com>
---
 drivers/oprofile/oprofile_stats.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/oprofile/oprofile_stats.c b/drivers/oprofile/oprofile_stats.c
index 59659ce..6fe7538 100644
--- a/drivers/oprofile/oprofile_stats.c
+++ b/drivers/oprofile/oprofile_stats.c
@@ -43,7 +43,7 @@  void oprofile_create_stats_files(struct dentry *root)
 	struct oprofile_cpu_buffer *cpu_buf;
 	struct dentry *cpudir;
 	struct dentry *dir;
-	char buf[10];
+	char buf[16];
 	int i;
 
 	dir = oprofilefs_mkdir(root, "stats");
@@ -52,7 +52,7 @@  void oprofile_create_stats_files(struct dentry *root)
 
 	for_each_possible_cpu(i) {
 		cpu_buf = &per_cpu(op_cpu_buffer, i);
-		snprintf(buf, 10, "cpu%d", i);
+		snprintf(buf, 16, "cpu%d", i);
 		cpudir = oprofilefs_mkdir(dir, buf);
 
 		/* Strictly speaking access to these ulongs is racy,
-- 
2.9.3