Skip to contentSkip to conversation

gcc-sanitizers: fix -Werror=maybe-uninitialized issue

PATCH Superseded [email protected] 3 participants 17 messages Mar 19, 2019 – Apr 22, 2019

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 gcc-sanitizers: fix -Werror=maybe-uninitialized issue

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);

Initialize st_mode member to fix the above issue.

---
 meta/recipes-devtools/gcc/gcc-8.3.inc              |  1 +
 .../0041-elf.c-initialize-st_mode-member.patch     | 33 ++++++++++++++++++++++
 2 files changed, 34 insertions(+)
 create mode 100644 meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch
meta/recipes-devtools/gcc/gcc-8.3.inc +1 -0
index f7bf257..3501325 100644
@@ -71,6 +71,7 @@ SRC_URI = "\
            file://0038-Re-introduce-spe-commandline-options.patch \
            file://0039-riscv-Disable-multilib-for-OE.patch \
            file://0040-powerpc-powerpc64-Add-support-for-musl-ldso.patch \
+           file://0041-elf.c-initialize-st_mode-member.patch \
 "
 SRC_URI[md5sum] = "65b210b4bfe7e060051f799e0f994896"
 SRC_URI[sha256sum] = "64baadfe6cc0f4947a84cb12d7f0dfaf45bb58b7e92461639596c21e02d97d2c"
meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch new file +33 -0
new file mode 100644
index 0000000..8e80c4e
@@ -0,0 +1,33 @@
+From 960f3e1f104192ceb80a97e665fadffda5285aab Mon Sep 17 00:00:00 2001
+From: Mingli Yu <[email protected]>
+Date: Tue, 19 Mar 2019 16:23:04 +0800
+Subject: [PATCH] elf.c: initialize st_mode member
+
+Initialize st_mode member to fix the below
+build failure when -Og included in compiler flag.
+| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
+| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+   return S_ISLNK (st.st_mode);
+
+Upstream-Status: Submitted[https://gcc.gnu.org/ml/gcc-patches/2019-03/msg00900.html]
+
+Signed-off-by: Mingli Yu <[email protected]>
+---
+ libbacktrace/elf.c | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/libbacktrace/elf.c b/libbacktrace/elf.c
+index f4863f0..561bdc2 100644
+--- a/libbacktrace/elf.c
++++ b/libbacktrace/elf.c
+@@ -766,6 +766,7 @@ static int
+ elf_is_symlink (const char *filename)
+ {
+   struct stat st;
++  st.st_mode = 0;
+ 
+   if (lstat (filename, &st) < 0)
+     return 0;
+-- 
+2.7.4
+
Signature · 1 line
2.7.4
author
#1 lore raw

Ping.

Thanks,

Quoted 69 lines · On 2019年03月19日 16:44, [email protected] wrote:

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);

Initialize st_mode member to fix the above issue.

---
  meta/recipes-devtools/gcc/gcc-8.3.inc              |  1 +
  .../0041-elf.c-initialize-st_mode-member.patch     | 33 ++++++++++++++++++++++
  2 files changed, 34 insertions(+)

create mode 100644 meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch

meta/recipes-devtools/gcc/gcc-8.3.inc +1 -0
index f7bf257..3501325 100644
@@ -71,6 +71,7 @@ SRC_URI = "\
             file://0038-Re-introduce-spe-commandline-options.patch \
             file://0039-riscv-Disable-multilib-for-OE.patch \
             file://0040-powerpc-powerpc64-Add-support-for-musl-ldso.patch \
+           file://0041-elf.c-initialize-st_mode-member.patch \
  "
  SRC_URI[md5sum] = "65b210b4bfe7e060051f799e0f994896"
  SRC_URI[sha256sum] = "64baadfe6cc0f4947a84cb12d7f0dfaf45bb58b7e92461639596c21e02d97d2c"
meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch new file +33 -0
new file mode 100644
index 0000000..8e80c4e
@@ -0,0 +1,33 @@
+From 960f3e1f104192ceb80a97e665fadffda5285aab Mon Sep 17 00:00:00 2001
+From: Mingli Yu <[email protected]>
+Date: Tue, 19 Mar 2019 16:23:04 +0800
+Subject: [PATCH] elf.c: initialize st_mode member
+
+Initialize st_mode member to fix the below
+build failure when -Og included in compiler flag.
+| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
+| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+   return S_ISLNK (st.st_mode);
+
+Upstream-Status: Submitted[https://gcc.gnu.org/ml/gcc-patches/2019-03/msg00900.html]
+
+Signed-off-by: Mingli Yu <[email protected]>
+---
+ libbacktrace/elf.c | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/libbacktrace/elf.c b/libbacktrace/elf.c
+index f4863f0..561bdc2 100644
+--- a/libbacktrace/elf.c
++++ b/libbacktrace/elf.c
+@@ -766,6 +766,7 @@ static int
+ elf_is_symlink (const char *filename)
+ {
+   struct stat st;
++  st.st_mode = 0;
+
+   if (lstat (filename, &st) < 0)
+     return 0;
+--
+2.7.4
+
author
#2 lore raw
In reply to [email protected]

Ping.

Any comments here?

Thanks,

Quoted 69 lines · On 2019年03月19日 16:44, [email protected] wrote:

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);

Initialize st_mode member to fix the above issue.

---
  meta/recipes-devtools/gcc/gcc-8.3.inc              |  1 +
  .../0041-elf.c-initialize-st_mode-member.patch     | 33 ++++++++++++++++++++++
  2 files changed, 34 insertions(+)

create mode 100644 meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch

meta/recipes-devtools/gcc/gcc-8.3.inc +1 -0
index f7bf257..3501325 100644
@@ -71,6 +71,7 @@ SRC_URI = "\
             file://0038-Re-introduce-spe-commandline-options.patch \
             file://0039-riscv-Disable-multilib-for-OE.patch \
             file://0040-powerpc-powerpc64-Add-support-for-musl-ldso.patch \
+           file://0041-elf.c-initialize-st_mode-member.patch \
  "
  SRC_URI[md5sum] = "65b210b4bfe7e060051f799e0f994896"
  SRC_URI[sha256sum] = "64baadfe6cc0f4947a84cb12d7f0dfaf45bb58b7e92461639596c21e02d97d2c"
meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch new file +33 -0
new file mode 100644
index 0000000..8e80c4e
@@ -0,0 +1,33 @@
+From 960f3e1f104192ceb80a97e665fadffda5285aab Mon Sep 17 00:00:00 2001
+From: Mingli Yu <[email protected]>
+Date: Tue, 19 Mar 2019 16:23:04 +0800
+Subject: [PATCH] elf.c: initialize st_mode member
+
+Initialize st_mode member to fix the below
+build failure when -Og included in compiler flag.
+| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
+| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+   return S_ISLNK (st.st_mode);
+
+Upstream-Status: Submitted[https://gcc.gnu.org/ml/gcc-patches/2019-03/msg00900.html]
+
+Signed-off-by: Mingli Yu <[email protected]>
+---
+ libbacktrace/elf.c | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/libbacktrace/elf.c b/libbacktrace/elf.c
+index f4863f0..561bdc2 100644
+--- a/libbacktrace/elf.c
++++ b/libbacktrace/elf.c
+@@ -766,6 +766,7 @@ static int
+ elf_is_symlink (const char *filename)
+ {
+   struct stat st;
++  st.st_mode = 0;
+
+   if (lstat (filename, &st) < 0)
+     return 0;
+--
+2.7.4
+
#3 lore raw
In reply to [email protected]
Quoted 53 lines · On Tue, Mar 19, 2019 at 1:45 AM <[email protected]> wrote:

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);

Initialize st_mode member to fix the above issue.

---
 meta/recipes-devtools/gcc/gcc-8.3.inc              |  1 +
 .../0041-elf.c-initialize-st_mode-member.patch     | 33 ++++++++++++++++++++++
 2 files changed, 34 insertions(+)
 create mode 100644 meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch
meta/recipes-devtools/gcc/gcc-8.3.inc +1 -0
index f7bf257..3501325 100644
@@ -71,6 +71,7 @@ SRC_URI = "\
            file://0038-Re-introduce-spe-commandline-options.patch \
            file://0039-riscv-Disable-multilib-for-OE.patch \
            file://0040-powerpc-powerpc64-Add-support-for-musl-ldso.patch \
+           file://0041-elf.c-initialize-st_mode-member.patch \
 "
 SRC_URI[md5sum] = "65b210b4bfe7e060051f799e0f994896"
 SRC_URI[sha256sum] = "64baadfe6cc0f4947a84cb12d7f0dfaf45bb58b7e92461639596c21e02d97d2c"
meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch new file +17 -0
new file mode 100644
index 0000000..8e80c4e
@@ -0,0 +1,33 @@
+From 960f3e1f104192ceb80a97e665fadffda5285aab Mon Sep 17 00:00:00 2001
+From: Mingli Yu <[email protected]>
+Date: Tue, 19 Mar 2019 16:23:04 +0800
+Subject: [PATCH] elf.c: initialize st_mode member
+
+Initialize st_mode member to fix the below
+build failure when -Og included in compiler flag.
+| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
+| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+   return S_ISLNK (st.st_mode);
+
+Upstream-Status: Submitted[https://gcc.gnu.org/ml/gcc-patches/2019-03/msg00900.html]
+
+Signed-off-by: Mingli Yu <[email protected]>
+---
+ libbacktrace/elf.c | 1 +
+ 1 file changed, 1 insertion(+)

meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch line 27

Quoted 10 lines
meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch new file +10 -0
+
+diff --git a/libbacktrace/elf.c b/libbacktrace/elf.c
+index f4863f0..561bdc2 100644
+--- a/libbacktrace/elf.c
++++ b/libbacktrace/elf.c
+@@ -766,6 +766,7 @@ static int
+ elf_is_symlink (const char *filename)
+ {
+   struct stat st;
++  st.st_mode = 0;

perhaps its safer to initialize whole struct to 0

struct stat st = {0};

Quoted 14 lines
meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-st_mode-member.patch new file +6 -1
+
+   if (lstat (filename, &st) < 0)
+     return 0;
+--
+2.7.4
+
--

2.7.4

--
_______________________________________________
Openembedded-core mailing list
[email protected]
http://lists.openembedded.org/mailman/listinfo/openembedded-core

#4 lore raw

PATCH v2 1/1 gcc-sanitizers: fix -Werror=maybe-uninitialized issue

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);

Initialize struct stat to fix the above issue.

---
 meta/recipes-devtools/gcc/gcc-8.3.inc              |  1 +
 .../0041-elf.c-initialize-struct-stat.patch        | 34 ++++++++++++++++++++++
 2 files changed, 35 insertions(+)
 create mode 100644 meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-struct-stat.patch
meta/recipes-devtools/gcc/gcc-8.3.inc +1 -0
index f7bf257..317c6af 100644
@@ -71,6 +71,7 @@ SRC_URI = "\
            file://0038-Re-introduce-spe-commandline-options.patch \
            file://0039-riscv-Disable-multilib-for-OE.patch \
            file://0040-powerpc-powerpc64-Add-support-for-musl-ldso.patch \
+           file://0041-elf.c-initialize-struct-stat.patch \
 "
 SRC_URI[md5sum] = "65b210b4bfe7e060051f799e0f994896"
 SRC_URI[sha256sum] = "64baadfe6cc0f4947a84cb12d7f0dfaf45bb58b7e92461639596c21e02d97d2c"
meta/recipes-devtools/gcc/gcc-8.3/0041-elf.c-initialize-struct-stat.patch new file +34 -0
new file mode 100644
index 0000000..56d62c9
@@ -0,0 +1,34 @@
+From eae259e4722a1c4606a1e6df7ee84a311a91f4a6 Mon Sep 17 00:00:00 2001
+From: Mingli Yu <[email protected]>
+Date: Mon, 8 Apr 2019 14:19:31 +0800
+Subject: [PATCH] elf.c: initialize struct stat
+
+Initialize struct stat to fix the below
+build failure when -Og included in compiler flag.
+| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
+| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+   return S_ISLNK (st.st_mode);
+
+Upstream-Status: Submitted[https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00261.html]
+
+Signed-off-by: Mingli Yu <[email protected]>
+---
+ libbacktrace/elf.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/libbacktrace/elf.c b/libbacktrace/elf.c
+index f4863f0..8c57047 100644
+--- a/libbacktrace/elf.c
++++ b/libbacktrace/elf.c
+@@ -765,7 +765,7 @@ elf_syminfo (struct backtrace_state *state, uintptr_t addr,
+ static int
+ elf_is_symlink (const char *filename)
+ {
+-  struct stat st;
++  struct stat st = {0};
+ 
+   if (lstat (filename, &st) < 0)
+     return 0;
+-- 
+2.7.4
+
Signature · 1 line
2.7.4
#5 lore raw

PATCH v3 1/1 gcc-sanitizers: fix -Werror=maybe-uninitialized issue

In reply to Khem Raj

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);

Per https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html,
the gcc upstream thinks the warning is a false positive and suggests to use -O2 rather than -Og
or -O1 when compiling that file, so pass -Wno-error to compiler when -Og is used to silence the error.

---
 meta/recipes-devtools/gcc/gcc-sanitizers.inc | 3 +++
 1 file changed, 3 insertions(+)
meta/recipes-devtools/gcc/gcc-sanitizers.inc +3 -0
index e5e8452..8b1d1c9 100644
@@ -45,6 +45,9 @@ INHIBIT_DEFAULT_DEPS = "1"
 ALLOW_EMPTY_${PN} = "1"
 DEPENDS = "gcc-runtime virtual/${TARGET_PREFIX}gcc"
 
+# used to fix ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+DEBUG_OPTIMIZATION_append = " -Wno-error"
+
 BBCLASSEXTEND = "nativesdk"
 
 PACKAGES = "${PN} ${PN}-dbg"
Signature · 1 line
2.7.4
#6 lore raw
Quoted 9 lines · On Mon, Apr 8, 2019 at 7:32 PM <[email protected]> wrote:

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);
Quoted 6 lines

Per https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html,
the gcc upstream thinks the warning is a false positive and suggests to use -O2 rather than -Og
or -O1 when compiling that file, so pass -Wno-error to compiler when -Og is used to silence the error.

Not particular to this change but in general if a package says that it does not support -Og
then we are just going to get into more and more untested grounds especially during runtime
so I wonder how useful it will be to use -Og for such packages or any other non supported
combination for that matter.

Quoted 22 lines
---
 meta/recipes-devtools/gcc/gcc-sanitizers.inc | 3 +++
 1 file changed, 3 insertions(+)
meta/recipes-devtools/gcc/gcc-sanitizers.inc +3 -1
index e5e8452..8b1d1c9 100644
@@ -45,6 +45,9 @@ INHIBIT_DEFAULT_DEPS = "1"
 ALLOW_EMPTY_${PN} = "1"
 DEPENDS = "gcc-runtime virtual/${TARGET_PREFIX}gcc"
 
+# used to fix ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+DEBUG_OPTIMIZATION_append = " -Wno-error"
+
 BBCLASSEXTEND = "nativesdk"
 
 PACKAGES = "${PN} ${PN}-dbg"
--

2.7.4

author
#7 lore raw
Quoted 17 lines · Khem Raj
Quoted 9 lines · On Mon, Apr 8, 2019 at 7:32 PM <[email protected]> wrote:

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);
Quoted 6 lines

Per https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html,
the gcc upstream thinks the warning is a false positive and suggests to use -O2 rather than -Og
or -O1 when compiling that file, so pass -Wno-error to compiler when -Og is used to silence the error.

Quoted 7 lines

Not particular to this change but in general if a package says that it does not support -Og
then we are just going to get into more and more untested grounds especially during runtime
so I wonder how useful it will be to use -Og for such packages or any other non supported
combination for that matter.

Hi Khem,

Any other suggestion to silence the gcc-sanitizers build error when DEBUG_BUILD is enabled?

Or just keep the logic DEBUG_OPTIMIZATION_append = " -Wno-error" as my patch stated?

Thanks,

Quoted 24 lines
Quoted 23 lines
---
  meta/recipes-devtools/gcc/gcc-sanitizers.inc | 3 +++
  1 file changed, 3 insertions(+)
meta/recipes-devtools/gcc/gcc-sanitizers.inc +3 -1
index e5e8452..8b1d1c9 100644
@@ -45,6 +45,9 @@ INHIBIT_DEFAULT_DEPS = "1"
  ALLOW_EMPTY_${PN} = "1"
  DEPENDS = "gcc-runtime virtual/${TARGET_PREFIX}gcc"
 
+# used to fix ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+DEBUG_OPTIMIZATION_append = " -Wno-error"
+
  BBCLASSEXTEND = "nativesdk"
 
  PACKAGES = "${PN} ${PN}-dbg"
--

2.7.4

#8 lore raw
In reply to Khem Raj
Quoted 17 lines · Khem Raj
Quoted 9 lines · On Mon, Apr 8, 2019 at 7:32 PM <[email protected]> wrote:

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);
Quoted 6 lines

Per https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html,
the gcc upstream thinks the warning is a false positive and suggests to use -O2 rather than -Og
or -O1 when compiling that file, so pass -Wno-error to compiler when -Og is used to silence the error.

Quoted 7 lines

Not particular to this change but in general if a package says that it does not support -Og
then we are just going to get into more and more untested grounds especially during runtime
so I wonder how useful it will be to use -Og for such packages or any other non supported
combination for that matter.

This has nothing to do with specific packages not supporting -Og or any other combination, it is just about how to best workaround a compiler bug temporarily.

gcc has bug(s) emitting bogus warnings with -Og, and these are build failures with some packages that build with -Werror.

None of this is related to whether or not the packages will work at runtime with -Og.

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed
#9 lore raw
Quoted 35 lines · Adrian Bunk
Quoted 7 lines · Khem Raj
Quoted 6 lines · On Mon, Apr 8, 2019 at 7:32 PM <[email protected]> wrote:

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":
|

./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:
In function 'elf_is_symlink':

Quoted 1 line
Quoted 1 line

|

../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21:
error: 'st.st_mode' may be used uninitialized in this function
[-Werror=maybe-uninitialized]

Quoted 9 lines
Quoted 8 lines

| return S_ISLNK (st.st_mode);

Per https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html,
the gcc upstream thinks the warning is a false positive and suggests to use -O2 rather than -Og
or -O1 when compiling that file, so pass -Wno-error to compiler when -Og is used to silence the error.

Quoted 7 lines

Not particular to this change but in general if a package says that it does not support -Og
then we are just going to get into more and more untested grounds especially during runtime
so I wonder how useful it will be to use -Og for such packages or any other non supported
combination for that matter.

This has nothing to do with specific packages not supporting -Og or any other combination, it is just about how to best workaround a compiler bug temporarily.

Quoted 5 lines

gcc has bug(s) emitting bogus warnings with -Og, and these are build failures with some packages that build with -Werror.

None of this is related to whether or not the packages will work at runtime with -Og.

What are you trying to convey ? That’s what I mentioned before I began my reply however to reiterate my point was if a package is not usually built and tested with this combination which is evident because it fails to build then how good would it be if we fix this error especially complex packages like compilers so is it worth to fix them or disable Og for them

Quoted 13 lines

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed

#10 lore raw
Quoted 6 lines · Khem Raj

What are you trying to convey ? That’s what I mentioned before I began my reply however to reiterate my point was if a package is not usually built and tested with this combination which is evident because it fails to build then how good would it be if we fix this error especially complex packages like compilers so is it worth to fix them or disable Og for them

Packages that usually get built and tested with -Og should be pretty rare, and these specific build failures are better at finding the rare packages that use -Werror than pointing at potential miscompilations.

From a distribution point of view, a package build with -Werror by default is arguably a bug since this frequently breaks when something is changed (usually the compiler version).

-Og is better suited than the -O that was previously used for debugging, but are we talking about debug builds or production builds? If users would be using DEBUG_OPTIMIZATION in production builds that would be wrong - this will always be a mostly untested situation with an increased probability of hitting bugs noone else has seen before.

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed
author
#11 lore raw
Quoted 12 lines · Adrian Bunk
Quoted 6 lines · Khem Raj

What are you trying to convey ? That’s what I mentioned before I began my reply however to reiterate my point was if a package is not usually built and tested with this combination which is evident because it fails to build then how good would it be if we fix this error especially complex packages like compilers so is it worth to fix them or disable Og for them

Packages that usually get built and tested with -Og should be pretty rare, and these specific build failures are better at finding the rare packages that use -Werror than pointing at potential miscompilations.

Quoted 7 lines
Quoted 1 line

From a distribution point of view, a package build with -Werror by

default is arguably a bug since this frequently breaks when something is changed (usually the compiler version).

-Og is better suited than the -O that was previously used for debugging, but are we talking about debug builds or production builds? If users would be using DEBUG_OPTIMIZATION in production builds that

Thanks Adrian and Khem's response!
We indeed don't use DEBUG_OPTIMIZATION in production build.

But still comes question: how to silence gcc-sanitizers build failure when debug build enabled?

Thanks,

Quoted 7 lines

would be wrong - this will always be a mostly untested situation with an increased probability of hitting bugs noone else has seen before.

cu
Adrian

#12 lore raw
Quoted 24 lines · Mingli
Quoted 12 lines · Adrian Bunk
Quoted 6 lines · Khem Raj

What are you trying to convey ? That’s what I mentioned before I began my reply however to reiterate my point was if a package is not usually built and tested with this combination which is evident because it fails to build then how good would it be if we fix this error especially complex packages like compilers so is it worth to fix them or disable Og for them

Packages that usually get built and tested with -Og should be pretty rare, and these specific build failures are better at finding the rare packages that use -Werror than pointing at potential miscompilations.

Quoted 7 lines
Quoted 1 line

From a distribution point of view, a package build with -Werror by

default is arguably a bug since this frequently breaks when something is changed (usually the compiler version).

-Og is better suited than the -O that was previously used for debugging, but are we talking about debug builds or production builds? If users would be using DEBUG_OPTIMIZATION in production builds that

Quoted 11 lines

Thanks Adrian and Khem's response!
We indeed don't use DEBUG_OPTIMIZATION in production build.

But still comes question: how to silence gcc-sanitizers build failure when debug build enabled?

Thanks,

Quoted 3 lines

would be wrong - this will always be a mostly untested situation with an increased probability of hitting bugs noone else has seen before.

disable warning as errors.

Quoted 4 lines
Quoted 4 lines

cu
Adrian

author
#13 lore raw
Quoted 31 lines · Khem Raj
Quoted 30 lines · Mingli
Quoted 12 lines · Adrian Bunk
Quoted 6 lines · Khem Raj

What are you trying to convey ? That’s what I mentioned before I began my reply however to reiterate my point was if a package is not usually built and tested with this combination which is evident because it fails to build then how good would it be if we fix this error especially complex packages like compilers so is it worth to fix them or disable Og for them

Packages that usually get built and tested with -Og should be pretty rare, and these specific build failures are better at finding the rare packages that use -Werror than pointing at potential miscompilations.

Quoted 7 lines
Quoted 1 line

From a distribution point of view, a package build with -Werror by

default is arguably a bug since this frequently breaks when something is changed (usually the compiler version).

-Og is better suited than the -O that was previously used for debugging, but are we talking about debug builds or production builds? If users would be using DEBUG_OPTIMIZATION in production builds that

Thanks Adrian and Khem's response!
We indeed don't use DEBUG_OPTIMIZATION in production build.

But still comes question: how to silence gcc-sanitizers build failure when debug build enabled?

Quoted 7 lines
Quoted 5 lines

Thanks,

Quoted 3 lines

would be wrong - this will always be a mostly untested situation with an increased probability of hitting bugs noone else has seen before.

disable warning as errors.

Hi Khem,

I don't quite get what you mean. Disable warning as errors?

As https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html, gcc upstream thinks the build error is meaningless when compile gcc-sanitizers with -Og and suggests not use -Og to compile for gcc-sanitizers.

So I send out the patch to add the workaround [DEBUG_OPTIMIZATION_append = " -Wno-error"] to silence the error when debug build is enabled for gcc-sanitizers. If the workaround isn't suitable, then how to silence the -Werror=maybe-uninitialized issue build error when debug build enabled?

Thanks,

Quoted 6 lines
Quoted 5 lines
Quoted 5 lines

cu
Adrian

#14 lore raw
Quoted 52 lines · Mingli
Quoted 31 lines · Khem Raj
Quoted 30 lines · Mingli
Quoted 12 lines · Adrian Bunk
Quoted 6 lines · Khem Raj

What are you trying to convey ? That’s what I mentioned before I began my reply however to reiterate my point was if a package is not usually built and tested with this combination which is evident because it fails to build then how good would it be if we fix this error especially complex packages like compilers so is it worth to fix them or disable Og for them

Packages that usually get built and tested with -Og should be pretty rare, and these specific build failures are better at finding the rare packages that use -Werror than pointing at potential miscompilations.

Quoted 7 lines
Quoted 1 line

From a distribution point of view, a package build with -Werror by

default is arguably a bug since this frequently breaks when something is changed (usually the compiler version).

-Og is better suited than the -O that was previously used for debugging, but are we talking about debug builds or production builds? If users would be using DEBUG_OPTIMIZATION in production builds that

Thanks Adrian and Khem's response!
We indeed don't use DEBUG_OPTIMIZATION in production build.

But still comes question: how to silence gcc-sanitizers build failure when debug build enabled?

Quoted 7 lines
Quoted 5 lines

Thanks,

Quoted 3 lines

would be wrong - this will always be a mostly untested situation with an increased probability of hitting bugs noone else has seen before.

disable warning as errors.

Hi Khem,

I don't quite get what you mean. Disable warning as errors?

As https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html, gcc upstream thinks the build error is meaningless when compile gcc-sanitizers with -Og and suggests not use -Og to compile for gcc-sanitizers.

Quoted 5 lines

So I send out the patch to add the workaround [DEBUG_OPTIMIZATION_append = " -Wno-error"] to silence the error when debug build is enabled for gcc-sanitizers.
If the workaround isn't suitable, then how to silence the -Werror=maybe-uninitialized issue build error when debug build enabled?

I was meaning to say that remove -Werror from cmdline completely

Quoted 9 lines

Thanks,

Quoted 6 lines
Quoted 5 lines
Quoted 5 lines

cu
Adrian

author
#15 lore raw
Quoted 53 lines · Khem Raj
Quoted 52 lines · Mingli
Quoted 31 lines · Khem Raj
Quoted 30 lines · Mingli
Quoted 12 lines · Adrian Bunk
Quoted 6 lines · Khem Raj

What are you trying to convey ? That’s what I mentioned before I began my reply however to reiterate my point was if a package is not usually built and tested with this combination which is evident because it fails to build then how good would it be if we fix this error especially complex packages like compilers so is it worth to fix them or disable Og for them

Packages that usually get built and tested with -Og should be pretty rare, and these specific build failures are better at finding the rare packages that use -Werror than pointing at potential miscompilations.

Quoted 7 lines
Quoted 1 line

From a distribution point of view, a package build with -Werror by

default is arguably a bug since this frequently breaks when something is changed (usually the compiler version).

-Og is better suited than the -O that was previously used for debugging, but are we talking about debug builds or production builds? If users would be using DEBUG_OPTIMIZATION in production builds that

Thanks Adrian and Khem's response!
We indeed don't use DEBUG_OPTIMIZATION in production build.

But still comes question: how to silence gcc-sanitizers build failure when debug build enabled?

Quoted 7 lines
Quoted 5 lines

Thanks,

Quoted 3 lines

would be wrong - this will always be a mostly untested situation with an increased probability of hitting bugs noone else has seen before.

disable warning as errors.

Hi Khem,

I don't quite get what you mean. Disable warning as errors?

As https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html, gcc upstream thinks the build error is meaningless when compile gcc-sanitizers with -Og and suggests not use -Og to compile for gcc-sanitizers.

Quoted 7 lines
Quoted 5 lines

So I send out the patch to add the workaround [DEBUG_OPTIMIZATION_append = " -Wno-error"] to silence the error when debug build is enabled for gcc-sanitizers.
If the workaround isn't suitable, then how to silence the -Werror=maybe-uninitialized issue build error when debug build enabled?

I was meaning to say that remove -Werror from cmdline completely

Thanks Khem! So this is what we did currently now in V3 patch.

Hi RP,
So any comments here/

Thanks,

Quoted 11 lines
Quoted 10 lines

Thanks,

Quoted 6 lines
Quoted 5 lines
Quoted 5 lines

cu
Adrian

#16 lore raw

PATCH v4 1/1 gcc-sanitizers: fix -Werror=maybe-uninitialized issue

In reply to Khem Raj

From: Mingli Yu <[email protected]>

When DEBUG_BUILD = "1" added in local.conf, there comes below build error when "bitbake gcc-sanitizers":

| ./../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c: In function 'elf_is_symlink':
| ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
|   return S_ISLNK (st.st_mode);

After commit[16643b0322 bitbake.conf: Use -Og in DEBUG_OPTIMIZATION] introduced, "-Og" added to compiler when debug
build enabled.

Per https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00315.html,
the gcc upstream thinks the warning is a false positive and suggests to use -O2 rather than -Og
or -O1 when compiling that file, so pass -Wno-error to compiler when -Og is used to silence the error.

---
 meta/recipes-devtools/gcc/gcc-sanitizers.inc | 3 +++
 1 file changed, 3 insertions(+)
meta/recipes-devtools/gcc/gcc-sanitizers.inc +3 -0
index e5e8452..8b1d1c9 100644
@@ -45,6 +45,9 @@ INHIBIT_DEFAULT_DEPS = "1"
 ALLOW_EMPTY_${PN} = "1"
 DEPENDS = "gcc-runtime virtual/${TARGET_PREFIX}gcc"
 
+# used to fix ../../../../../../../../../work-shared/gcc-8.3.0-r0/gcc-8.3.0/libsanitizer/libbacktrace/../../libbacktrace/elf.c:772:21: error: 'st.st_mode' may be used uninitialized in this function [-Werror=maybe-uninitialized]
+DEBUG_OPTIMIZATION_append = " -Wno-error"
+
 BBCLASSEXTEND = "nativesdk"
 
 PACKAGES = "${PN} ${PN}-dbg"
Signature · 1 line
2.7.4