This is the mail archive of the
gdb-cvs@sourceware.org
mailing list for the GDB project.
[binutils-gdb] Fix build breakage on GNU/Linux AArch64, take 2
- From: Sergio Durigan Junior <sergiodj at sourceware dot org>
- To: gdb-cvs at sourceware dot org
- Date: 10 Sep 2017 22:13:01 -0000
- Subject: [binutils-gdb] Fix build breakage on GNU/Linux AArch64, take 2
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=a059f00c857d088a7fe55d2dbded9d3210af4989
commit a059f00c857d088a7fe55d2dbded9d3210af4989
Author: Sergio Durigan Junior <sergiodj@redhat.com>
Date: Sun Sep 10 17:50:29 2017 -0400
Fix build breakage on GNU/Linux AArch64, take 2
The last commit unfortunately was not enough to fix the build breakage
on AArch64. I made a mistake and did not test it alone on BuildBot,
but along with another patch that was responsible for fixing the
breakage.
The failure is:
In file included from /usr/include/string.h:640:0,
from build-gnulib-gdbserver/import/string.h:41,
from ../../../binutils-gdb/gdb/gdbserver/../common/common-defs.h:56,
from ../../../binutils-gdb/gdb/gdbserver/server.h:22,
from ../../../binutils-gdb/gdb/gdbserver/regcache.c:19:
In function â??void* memset(void*, int, size_t)â??,
inlined from â??regcache* init_register_cache(regcache*, const target_desc*, unsigned char*)â?? at ../../../binutils-gdb/gdb/gdbserver/regcache.c:150:50:
/usr/include/aarch64-linux-gnu/bits/string3.h:81:32: error: call to â??__warn_memset_zero_lenâ?? declared with attribute warning: memset used with constant zero length parameter; this could be due to transposed parameters [-Werror]
__warn_memset_zero_len ();
^
In function â??void* memset(void*, int, size_t)â??,
inlined from â??regcache* get_thread_regcache(thread_info*, int)â?? at ../../../binutils-gdb/gdb/gdbserver/regcache.c:57:60:
/usr/include/aarch64-linux-gnu/bits/string3.h:81:32: error: call to â??__warn_memset_zero_lenâ?? declared with attribute warning: memset used with constant zero length parameter; this could be due to transposed parameters [-Werror]
__warn_memset_zero_len ();
This is likely due to a GCC bug, because for some reason the compiler
assumes that the third argument to the memset:
memset (regcache->register_status, REG_UNAVAILABLE,
VEC_length (tdesc_reg_p, regcache->tdesc->reg_defs));
is always zero, which is not always true.
Anyway, the simple fix for this is to guard the memset calls with:
if (!VEC_empty (tdesc_reg_p, regcache->tdesc->reg_defs))
This time, I made sure to regtest only this patch on BuildBot, and it
finally solved the breakage.
gdb/gdbserver/ChangeLog:
2017-09-10 Sergio Durigan Junior <sergiodj@redhat.com>
* regcache.c (get_thread_regcache): Guard calls to "memset"
with "!VEC_empty".
Diff:
---
gdb/gdbserver/ChangeLog | 5 +++++
gdb/gdbserver/regcache.c | 10 ++++++----
2 files changed, 11 insertions(+), 4 deletions(-)
diff --git a/gdb/gdbserver/ChangeLog b/gdb/gdbserver/ChangeLog
index 345abb3..d9f80c9 100644
--- a/gdb/gdbserver/ChangeLog
+++ b/gdb/gdbserver/ChangeLog
@@ -1,5 +1,10 @@
2017-09-10 Sergio Durigan Junior <sergiodj@redhat.com>
+ * regcache.c (get_thread_regcache): Guard calls to "memset"
+ with "!VEC_empty".
+
+2017-09-10 Sergio Durigan Junior <sergiodj@redhat.com>
+
* linux-low.c (handle_extended_wait): Use
"allocate_target_description" instead of "XNEW".
* linux-x86-low.c (initialize_low_arch): Likewise.
diff --git a/gdb/gdbserver/regcache.c b/gdb/gdbserver/regcache.c
index 43e78a5..d1a534c 100644
--- a/gdb/gdbserver/regcache.c
+++ b/gdb/gdbserver/regcache.c
@@ -53,8 +53,9 @@ get_thread_regcache (struct thread_info *thread, int fetch)
current_thread = thread;
/* Invalidate all registers, to prevent stale left-overs. */
- memset (regcache->register_status, REG_UNAVAILABLE,
- VEC_length (tdesc_reg_p, regcache->tdesc->reg_defs));
+ if (!VEC_empty (tdesc_reg_p, regcache->tdesc->reg_defs))
+ memset (regcache->register_status, REG_UNAVAILABLE,
+ VEC_length (tdesc_reg_p, regcache->tdesc->reg_defs));
fetch_inferior_registers (regcache, -1);
current_thread = saved_thread;
regcache->registers_valid = 1;
@@ -146,8 +147,9 @@ init_register_cache (struct regcache *regcache,
regcache->registers_owned = 1;
regcache->register_status
= (unsigned char *) xmalloc (VEC_length (tdesc_reg_p, tdesc->reg_defs));
- memset ((void *) regcache->register_status, REG_UNAVAILABLE,
- VEC_length (tdesc_reg_p, tdesc->reg_defs));
+ if (!VEC_empty (tdesc_reg_p, tdesc->reg_defs))
+ memset ((void *) regcache->register_status, REG_UNAVAILABLE,
+ VEC_length (tdesc_reg_p, tdesc->reg_defs));
#else
gdb_assert_not_reached ("can't allocate memory from the heap");
#endif