From: Ben Hutchings <ben@decadent.org.uk>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk,
Steven Rostedt <srostedt@redhat.com>,
Steven Rostedt <rostedt@goodmis.org>
Subject: [ 36/49] ftrace: Do not function trace inlined functions
Date: Sun, 13 Jan 2013 17:43:31 +0000 [thread overview]
Message-ID: <20130113174303.320089125@decadent.org.uk> (raw)
In-Reply-To: <20130113174255.736888844@decadent.org.uk>
3.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <srostedt@redhat.com>
commit 45959ee7aa645815a5ce303a0ea1e48a21e67c6a upstream.
When gcc inlines a function, it does not mark it with the mcount
prologue, which in turn means that inlined functions are not traced
by the function tracer. But if CONFIG_OPTIMIZE_INLINING is set, then
gcc is allowed not to inline a function that is marked inline.
Depending on the options and the compiler, a function may or may
not be traced by the function tracer, depending on whether gcc
decides to inline a function or not. This has caused several
problems in the pass becaues gcc is not always consistent with
what it decides to inline between different gcc versions.
Some places should not be traced (like paravirt native_* functions)
and these are mostly marked as inline. When gcc decides not to
inline the function, and if that function should not be traced, then
the ftrace function tracer will suddenly break when it use to work
fine. This becomes even harder to debug when different versions of
gcc will not inline that function, making the same kernel and config
work for some gcc versions and not work for others.
By making all functions marked inline to not be traced will remove
the ambiguity that gcc adds when it comes to tracing functions marked
inline. All gcc versions will be consistent with what functions are
traced and having volatile working code will be removed.
Note, only the inline macro when CONFIG_OPTIMIZE_INLINING is set needs
to have notrace added, as the attribute __always_inline will force
the function to be inlined and then not traced.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
include/linux/compiler-gcc.h | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h
index 59e4028..3fd17c2 100644
--- a/include/linux/compiler-gcc.h
+++ b/include/linux/compiler-gcc.h
@@ -50,6 +50,11 @@
# define inline inline __attribute__((always_inline))
# define __inline__ __inline__ __attribute__((always_inline))
# define __inline __inline __attribute__((always_inline))
+#else
+/* A lot of inline functions can cause havoc with function tracing */
+# define inline inline notrace
+# define __inline__ __inline__ notrace
+# define __inline __inline notrace
#endif
#define __deprecated __attribute__((deprecated))
next prev parent reply other threads:[~2013-01-13 17:51 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-13 17:42 [ 00/49] 3.2.37-stable review Ben Hutchings
2013-01-13 17:42 ` [ 01/49] ext4: fix extent tree corruption caused by hole punch Ben Hutchings
2013-01-13 17:42 ` [ 02/49] i915: ensure that VGA plane is disabled Ben Hutchings
2013-01-13 17:42 ` [ 03/49] ext4: check dioread_nolock on remount Ben Hutchings
2013-01-13 17:42 ` [ 04/49] jbd2: fix assertion failure in jbd2_journal_flush() Ben Hutchings
2013-01-13 17:43 ` [ 05/49] hwmon: (lm73} Detect and report i2c bus errors Ben Hutchings
2013-01-13 17:43 ` [ 06/49] ext4: do not try to write superblock on ro remount w/o journal Ben Hutchings
2013-01-13 17:43 ` [ 07/49] PCI: Reduce Ricoh 0xe822 SD card reader base clock frequency to 50MHz Ben Hutchings
2013-01-13 17:43 ` [ 08/49] mm: Fix PageHead when !CONFIG_PAGEFLAGS_EXTENDED Ben Hutchings
2013-01-13 17:43 ` [ 09/49] cifs: adjust sequence number downward after signing NT_CANCEL request Ben Hutchings
2013-01-13 17:43 ` [ 10/49] tmpfs mempolicy: fix /proc/mounts corrupting memory Ben Hutchings
2013-01-13 17:43 ` [ 11/49] p54usb: add USB ID for T-Com Sinus 154 data II Ben Hutchings
2013-01-13 17:43 ` [ 12/49] ath9k_hw: Fix RX gain initvals for AR9485 Ben Hutchings
2013-01-13 17:43 ` [ 13/49] p54usb: add USBIDs for two more p54usb devices Ben Hutchings
2013-01-13 17:43 ` [ 14/49] powerpc/vdso: Remove redundant locking in update_vsyscall_tz() Ben Hutchings
2013-01-13 17:43 ` [ 15/49] powerpc: Add missing NULL terminator to avoid boot panic on PPC40x Ben Hutchings
2013-01-13 17:43 ` [ 16/49] drm/radeon: add connector table for SAM440ep embedded board Ben Hutchings
2013-01-13 17:43 ` [ 17/49] drm/radeon: add connector table for Mac G4 Silver Ben Hutchings
2013-01-14 18:48 ` Albrecht Dreß
2013-01-13 17:43 ` [ 18/49] drm/radeon: Properly handle DDC probe for DP bridges Ben Hutchings
2013-01-13 17:43 ` [ 19/49] NFSv4: Add ACCESS operation to OPEN compound Ben Hutchings
2013-01-15 18:42 ` Herton Ronaldo Krzesinski
2013-01-16 1:08 ` Ben Hutchings
2013-01-13 17:43 ` [ 20/49] NFSv4: dont check MAY_WRITE access bit in OPEN Ben Hutchings
2013-01-13 17:43 ` [ 21/49] NFS4: nfs4_opendata_access should return errno Ben Hutchings
2013-01-13 17:43 ` [ 22/49] NFS: Fix access to suid/sgid executables Ben Hutchings
2013-01-13 17:43 ` [ 23/49] drm/nouveau: fix init with agpgart-uninorth Ben Hutchings
2013-01-13 17:43 ` [ 24/49] video: mxsfb: fix crash when unblanking the display Ben Hutchings
2013-01-13 17:43 ` [ 25/49] nfs: fix null checking in nfs_get_option_str() Ben Hutchings
2013-01-13 17:43 ` [ 26/49] SUNRPC: Ensure that we free the rpc_task after cleanups are done Ben Hutchings
2013-01-13 17:43 ` [ 27/49] ACPI / scan: Do not use dummy HID for system bus ACPI nodes Ben Hutchings
2013-01-13 17:43 ` [ 28/49] mm: limit mmu_gather batching to fix soft lockups on !CONFIG_PREEMPT Ben Hutchings
2013-01-13 17:43 ` [ 29/49] drivers/rtc/rtc-vt8500.c: correct handling of CR_24H bitfield Ben Hutchings
2013-01-13 17:43 ` [ 30/49] drivers/rtc/rtc-vt8500.c: fix handling of data passed in struct rtc_time Ben Hutchings
2013-01-13 17:43 ` [ 31/49] udf: dont increment lenExtents while writing to a hole Ben Hutchings
2013-01-13 17:43 ` [ 32/49] epoll: prevent missed events on EPOLL_CTL_MOD Ben Hutchings
2013-01-13 17:43 ` [ 33/49] rt2x00: Dont let mac80211 send a BAR when an AMPDU subframe fails Ben Hutchings
2013-01-13 17:43 ` [ 34/49] mac80211: introduce IEEE80211_HW_TEARDOWN_AGGR_ON_BAR_FAIL Ben Hutchings
2013-01-13 17:43 ` [ 35/49] Revert: "rt2x00: Dont let mac80211 send a BAR when an AMPDU subframe fails" Ben Hutchings
2013-01-13 17:43 ` Ben Hutchings [this message]
2013-01-13 17:43 ` [ 37/49] sparc: huge_ptep_set_* functions need to call set_huge_pte_at() Ben Hutchings
2013-01-13 17:43 ` [ 38/49] inet: Fix kmemleak in tcp_v4/6_syn_recv_sock and dccp_v4/6_request_recv_sock Ben Hutchings
2013-01-13 17:43 ` [ 39/49] net: sched: integer overflow fix Ben Hutchings
2013-01-13 17:43 ` [ 40/49] tcp: implement RFC 5961 3.2 Ben Hutchings
2013-01-13 17:43 ` [ 41/49] tcp: implement RFC 5961 4.2 Ben Hutchings
2013-01-13 17:43 ` [ 42/49] tcp: refine SYN handling in tcp_validate_incoming Ben Hutchings
2013-01-13 17:43 ` [ 43/49] tcp: tcp_replace_ts_recent() should not be called from tcp_validate_incoming() Ben Hutchings
2013-01-13 17:43 ` [ 44/49] tcp: RFC 5961 5.2 Blind Data Injection Attack Mitigation Ben Hutchings
2013-01-13 17:43 ` [ 45/49] [SCSI] mvsas: Fix oops when ata commond timeout Ben Hutchings
2013-01-13 17:43 ` [ 46/49] RDMA/nes: Fix for crash when registering zero length MR for CQ Ben Hutchings
2013-01-13 17:43 ` [ 47/49] RDMA/nes: Fix for terminate timer crash Ben Hutchings
2013-01-13 17:43 ` [ 48/49] ACPI : do not use Lid and Sleep button for S5 wakeup Ben Hutchings
2013-01-13 17:43 ` [ 49/49] aoe: do not call bdi_init after blk_alloc_queue Ben Hutchings
2013-01-13 22:44 ` [ 00/49] 3.2.37-stable review Ben Hutchings
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=20130113174303.320089125@decadent.org.uk \
--to=ben@decadent.org.uk \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=linux-kernel@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=srostedt@redhat.com \
--cc=stable@vger.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox