2016-04-05 18:29:07

by Suman Anna

[permalink] [raw]
Subject: [PATCH] clk: ti: dflt: remove redundant unlikely

Commit 7aba4f5201d1 ("clk: ti: dflt: fix enable_reg validity check")
fixed a validation check by using an IS_ERR() macro within the
existing unlikely expression, but IS_ERR() macro already has an
unlikely inside it, so get rid of the redundant unlikely macro
from the validation check.

Reported-by: Stephen Boyd <[email protected]>
Signed-off-by: Suman Anna <[email protected]>
---
drivers/clk/ti/clkt_dflt.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clk/ti/clkt_dflt.c b/drivers/clk/ti/clkt_dflt.c
index 1ddc288fce4e..c6ae563801d7 100644
--- a/drivers/clk/ti/clkt_dflt.c
+++ b/drivers/clk/ti/clkt_dflt.c
@@ -222,7 +222,7 @@ int omap2_dflt_clk_enable(struct clk_hw *hw)
}
}

- if (unlikely(IS_ERR(clk->enable_reg))) {
+ if (IS_ERR(clk->enable_reg)) {
pr_err("%s: %s missing enable_reg\n", __func__,
clk_hw_get_name(hw));
ret = -EINVAL;
--
2.7.4


2016-04-05 21:30:21

by Joe Perches

[permalink] [raw]
Subject: Re: [PATCH] clk: ti: dflt: remove redundant unlikely

On Tue, 2016-04-05 at 13:28 -0500, Suman Anna wrote:
> Commit 7aba4f5201d1 ("clk: ti: dflt: fix enable_reg validity check")
> fixed a validation check by using an IS_ERR() macro within the
> existing unlikely expression, but IS_ERR() macro already has an
> unlikely inside it, so get rid of the redundant unlikely macro
> from the validation check.
[]
> diff --git a/drivers/clk/ti/clkt_dflt.c b/drivers/clk/ti/clkt_dflt.c
[]
> @@ -222,7 +222,7 @@ int omap2_dflt_clk_enable(struct clk_hw *hw)
> ? }
> ? }
> ?
> - if (unlikely(IS_ERR(clk->enable_reg))) {
> + if (IS_ERR(clk->enable_reg)) {
> ? pr_err("%s: %s missing enable_reg\n", __func__,
> ? ???????clk_hw_get_name(hw));
> ? ret = -EINVAL;

There are several of these:

$ git grep -n -E "likely.*\bIS_ERR\s*\(" *
drivers/clk/ti/clkt_dflt.c:225: if (unlikely(IS_ERR(clk->enable_reg))) {
drivers/md/dm-verity-fec.c:76:??if (unlikely(IS_ERR(res))) {
drivers/md/dm-verity-fec.c:168:?????????????????if (unlikely(IS_ERR(par)))
drivers/md/dm-verity-fec.c:253:?????????if (unlikely(IS_ERR(bbuf))) {
drivers/net/ethernet/broadcom/bnxt/bnxt.c:1109:?????????if (unlikely(IS_ERR(skb)))
drivers/net/ethernet/hisilicon/hns/hns_enet.c:1010:?????if (unlikely(!phy_dev) || IS_ERR(phy_dev))
drivers/net/ethernet/sun/ldmvsw.c:302:??if (unlikely(IS_ERR(vp))) {
drivers/scsi/bnx2fc/bnx2fc_fcoe.c:2547: if (likely(!IS_ERR(thread))) {
drivers/scsi/bnx2i/bnx2i_init.c:431:????if (likely(!IS_ERR(thread))) {
drivers/scsi/fcoe/fcoe.c:1263:??if (likely(!IS_ERR(thread))) {
drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd.c:1500:???????????????if (likely(!IS_ERR(pfmr))) {
drivers/thermal/intel_powerclamp.c:541:?????????if (likely(!IS_ERR(thread))) {
drivers/thermal/intel_powerclamp.c:591:?????????if (likely(!IS_ERR(thread))) {
drivers/tty/serial/serial_core.c:2682:??if (likely(!IS_ERR(tty_dev))) {
fs/gfs2/dir.c:854:??????if (unlikely(dent == NULL || IS_ERR(dent))) {
fs/ntfs/inode.c:2724:???if (unlikely(err || IS_ERR(m))) {
fs/ntfs/lcnalloc.c:744: if (likely(page && !IS_ERR(page))) {
fs/ntfs/mft.c:85:???????if (likely(!IS_ERR(page))) {
fs/ntfs/mft.c:168:??????if (likely(!IS_ERR(m)))
fs/ntfs/mft.c:285:??????????????if (likely(!IS_ERR(m))) {
fs/ntfs/mft.c:1311:?????if (unlikely(IS_ERR(rl) || !rl->length || rl->lcn < 0)) {
fs/ntfs/mft.c:1742:?????if (unlikely(IS_ERR(rl) || !rl->length || rl->lcn < 0)) {
fs/ntfs/mft.c:1784:?????????????if (likely(!IS_ERR(rl2)))
fs/ntfs/namei.c:132:????????????if (likely(!IS_ERR(dent_inode))) {
fs/ntfs/runlist.c:968:??if (likely(!IS_ERR(old_rl)))
fs/ntfs/super.c:1492:???if (unlikely(IS_ERR(tmp_ino) || is_bad_inode(tmp_ino))) {
net/ipv6/addrconf.c:3212:???????????????if (unlikely(IS_ERR(rt)))
net/kcm/kcmsock.c:1969: if (unlikely(IS_ERR(newfile))) {
net/openvswitch/datapath.c:1315:????????????????if (likely(!IS_ERR(reply))) {
net/socket.c:399:???????if (likely(!IS_ERR(newfile))) {

Maybe a new checkpatch test like:
---
?scripts/checkpatch.pl | 7 +++++++
?1 file changed, 7 insertions(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index e3d9c34..25e81b9 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -5734,6 +5734,13 @@ sub process {
? }
? }
?
+# check for IS_ERR with likely
+
+ if ($line =~ /\b((?:un)?likely)\s*\(\s*\!?\s*IS_ERR\s*\(/) {
+ WARN("IS_ERR_IS_UNLIKELY",
+ ?????"Using $1 isn't recommended as IS_ERR already uses unlikely\n" . $herecurr);
+ }
+
?# check for semaphores initialized locked
? if ($line =~ /^.\s*sema_init.+,\W?0\W?\)/) {
? WARN("CONSIDER_COMPLETION",

2016-04-15 22:52:53

by Stephen Boyd

[permalink] [raw]
Subject: Re: [PATCH] clk: ti: dflt: remove redundant unlikely

On 04/05, Suman Anna wrote:
> Commit 7aba4f5201d1 ("clk: ti: dflt: fix enable_reg validity check")
> fixed a validation check by using an IS_ERR() macro within the
> existing unlikely expression, but IS_ERR() macro already has an
> unlikely inside it, so get rid of the redundant unlikely macro
> from the validation check.
>
> Reported-by: Stephen Boyd <[email protected]>
> Signed-off-by: Suman Anna <[email protected]>
> ---

Applied to clk-next

--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

2016-04-15 22:53:25

by Stephen Boyd

[permalink] [raw]
Subject: Re: [PATCH] clk: ti: dflt: remove redundant unlikely

On 04/05, Joe Perches wrote:
>
> There are several of these:
[...]
>
> Maybe a new checkpatch test like:

Looks good to me.
--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project