2022-01-05 19:34:56

by Hugo Villeneuve

[permalink] [raw]
Subject: [PATCH] rtc: isl1208: avoid unnecessary rc variable tests

From: Hugo Villeneuve <[email protected]>

The rc variable doesn't need to be tested a second time when the <if> block
evaluates to false.

Signed-off-by: Hugo Villeneuve <[email protected]>
---
drivers/rtc/rtc-isl1208.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/rtc/rtc-isl1208.c b/drivers/rtc/rtc-isl1208.c
index 182dfa605515..c7f04df5a0b6 100644
--- a/drivers/rtc/rtc-isl1208.c
+++ b/drivers/rtc/rtc-isl1208.c
@@ -880,15 +880,17 @@ isl1208_probe(struct i2c_client *client, const struct i2c_device_id *id)
if (rc)
return rc;

- if (client->irq > 0)
+ if (client->irq > 0) {
rc = isl1208_setup_irq(client, client->irq);
- if (rc)
- return rc;
+ if (rc)
+ return rc;
+ }

- if (evdet_irq > 0 && evdet_irq != client->irq)
+ if (evdet_irq > 0 && evdet_irq != client->irq) {
rc = isl1208_setup_irq(client, evdet_irq);
- if (rc)
- return rc;
+ if (rc)
+ return rc;
+ }

rc = devm_rtc_nvmem_register(isl1208->rtc, &isl1208->nvmem_config);
if (rc)
--
2.30.2



2022-01-05 20:01:20

by Alexandre Belloni

[permalink] [raw]
Subject: Re: [PATCH] rtc: isl1208: avoid unnecessary rc variable tests

On 05/01/2022 14:34:39-0500, Hugo Villeneuve wrote:
> From: Hugo Villeneuve <[email protected]>
>
> The rc variable doesn't need to be tested a second time when the <if> block
> evaluates to false.
>

rc is not tested a second time, here is the relevant listing:

- if (client->irq > 0)
+ if (client->irq > 0) {
ffffffff81aef647: 41 8b b5 bc 01 00 00 mov 0x1bc(%r13),%esi
ffffffff81aef64e: 85 f6 test %esi,%esi
ffffffff81aef650: 0f 8f 35 01 00 00 jg ffffffff81aef78b <isl1208_probe+0x314>
rc = isl1208_setup_irq(client, client->irq);
if (rc)
return rc;
+ }

- if (evdet_irq > 0 && evdet_irq != client->irq)
+ if (evdet_irq > 0 && evdet_irq != client->irq) {
ffffffff81aef656: 85 db test %ebx,%ebx
ffffffff81aef658: 7e 0d jle ffffffff81aef667 <isl1208_probe+0x1f0>
ffffffff81aef65a: 41 39 9d bc 01 00 00 cmp %ebx,0x1bc(%r13)
@@ -1663,6 +1664,7 @@ ffffffff81aef661: 0f 85 0a 01 00 00
rc = isl1208_setup_irq(client, evdet_irq);
if (rc)
return rc;
+ }

As you can see, no change in assembly but it is worse to read. gcc on
arm behaves the same way.

> Signed-off-by: Hugo Villeneuve <[email protected]>
> ---
> drivers/rtc/rtc-isl1208.c | 14 ++++++++------
> 1 file changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/rtc/rtc-isl1208.c b/drivers/rtc/rtc-isl1208.c
> index 182dfa605515..c7f04df5a0b6 100644
> --- a/drivers/rtc/rtc-isl1208.c
> +++ b/drivers/rtc/rtc-isl1208.c
> @@ -880,15 +880,17 @@ isl1208_probe(struct i2c_client *client, const struct i2c_device_id *id)
> if (rc)
> return rc;
>
> - if (client->irq > 0)
> + if (client->irq > 0) {
> rc = isl1208_setup_irq(client, client->irq);
> - if (rc)
> - return rc;
> + if (rc)
> + return rc;
> + }
>
> - if (evdet_irq > 0 && evdet_irq != client->irq)
> + if (evdet_irq > 0 && evdet_irq != client->irq) {
> rc = isl1208_setup_irq(client, evdet_irq);
> - if (rc)
> - return rc;
> + if (rc)
> + return rc;
> + }
>
> rc = devm_rtc_nvmem_register(isl1208->rtc, &isl1208->nvmem_config);
> if (rc)
> --
> 2.30.2
>

--
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

2022-01-05 20:35:04

by Hugo Villeneuve

[permalink] [raw]
Subject: Re: [PATCH] rtc: isl1208: avoid unnecessary rc variable tests

On Wed, 5 Jan 2022 21:01:10 +0100
Alexandre Belloni <[email protected]> wrote:

> On 05/01/2022 14:34:39-0500, Hugo Villeneuve wrote:
> > From: Hugo Villeneuve <[email protected]>
> >
> > The rc variable doesn't need to be tested a second time when the <if> block
> > evaluates to false.
> >
>
> rc is not tested a second time, here is the relevant listing:
>
> - if (client->irq > 0)
> + if (client->irq > 0) {
> ffffffff81aef647: 41 8b b5 bc 01 00 00 mov 0x1bc(%r13),%esi
> ffffffff81aef64e: 85 f6 test %esi,%esi
> ffffffff81aef650: 0f 8f 35 01 00 00 jg ffffffff81aef78b <isl1208_probe+0x314>
> rc = isl1208_setup_irq(client, client->irq);
> if (rc)
> return rc;
> + }
>
> - if (evdet_irq > 0 && evdet_irq != client->irq)
> + if (evdet_irq > 0 && evdet_irq != client->irq) {
> ffffffff81aef656: 85 db test %ebx,%ebx
> ffffffff81aef658: 7e 0d jle ffffffff81aef667 <isl1208_probe+0x1f0>
> ffffffff81aef65a: 41 39 9d bc 01 00 00 cmp %ebx,0x1bc(%r13)
> @@ -1663,6 +1664,7 @@ ffffffff81aef661: 0f 85 0a 01 00 00
> rc = isl1208_setup_irq(client, evdet_irq);
> if (rc)
> return rc;
> + }
>
> As you can see, no change in assembly but it is worse to read. gcc on
> arm behaves the same way.

Hi Alexandre,
I am not sure that I fully understand your assembly code analysis. Maybe my patch comment was misleading, because I am not talking about a redundant test inside the if block, but ouside of it (after it).

Here is the original code with my annotations. Let's assume that the variable client->irq = 0:

---
/* If client->irq = 0, then function isl1208_setup_irq() will not be called, and rc will not be modified: */
if (client->irq > 0)
rc = isl1208_setup_irq(client, client->irq);

/* If rc hasn't been modified, there is no need to re-test its value here: */
if (rc)
return rc;
---

After the patch, this code section becomes:

---
if (client->irq > 0) {
rc = isl1208_setup_irq(client, client->irq);
if (rc)
return rc;
}
---

For me it is more logical and clearer like this. Moreover, you can see that at line 873 of the original driver, the same kind of mechanism is used:

---
if (isl1208->config->has_timestamp) {
rc = rtc_add_group(isl1208->rtc, &isl1219_rtc_sysfs_files);
if (rc)
return rc;
}
---

Regards,
Hugo.


> > Signed-off-by: Hugo Villeneuve <[email protected]>
> > ---
> > drivers/rtc/rtc-isl1208.c | 14 ++++++++------
> > 1 file changed, 8 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/rtc/rtc-isl1208.c b/drivers/rtc/rtc-isl1208.c
> > index 182dfa605515..c7f04df5a0b6 100644
> > --- a/drivers/rtc/rtc-isl1208.c
> > +++ b/drivers/rtc/rtc-isl1208.c
> > @@ -880,15 +880,17 @@ isl1208_probe(struct i2c_client *client, const struct i2c_device_id *id)
> > if (rc)
> > return rc;
> >
> > - if (client->irq > 0)
> > + if (client->irq > 0) {
> > rc = isl1208_setup_irq(client, client->irq);
> > - if (rc)
> > - return rc;
> > + if (rc)
> > + return rc;
> > + }
> >
> > - if (evdet_irq > 0 && evdet_irq != client->irq)
> > + if (evdet_irq > 0 && evdet_irq != client->irq) {
> > rc = isl1208_setup_irq(client, evdet_irq);
> > - if (rc)
> > - return rc;
> > + if (rc)
> > + return rc;
> > + }
> >
> > rc = devm_rtc_nvmem_register(isl1208->rtc, &isl1208->nvmem_config);
> > if (rc)
> > --
> > 2.30.2
> >
>
> --
> Alexandre Belloni, co-owner and COO, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com
>


--
Hugo Villeneuve <[email protected]>

2022-01-05 22:26:03

by Alexandre Belloni

[permalink] [raw]
Subject: Re: [PATCH] rtc: isl1208: avoid unnecessary rc variable tests

On 05/01/2022 15:34:46-0500, Hugo Villeneuve wrote:
> On Wed, 5 Jan 2022 21:01:10 +0100
> Alexandre Belloni <[email protected]> wrote:
>
> > On 05/01/2022 14:34:39-0500, Hugo Villeneuve wrote:
> > > From: Hugo Villeneuve <[email protected]>
> > >
> > > The rc variable doesn't need to be tested a second time when the <if> block
> > > evaluates to false.
> > >
> >
> > rc is not tested a second time, here is the relevant listing:
> >
> > - if (client->irq > 0)
> > + if (client->irq > 0) {
> > ffffffff81aef647: 41 8b b5 bc 01 00 00 mov 0x1bc(%r13),%esi
> > ffffffff81aef64e: 85 f6 test %esi,%esi
> > ffffffff81aef650: 0f 8f 35 01 00 00 jg ffffffff81aef78b <isl1208_probe+0x314>
> > rc = isl1208_setup_irq(client, client->irq);
> > if (rc)
> > return rc;
> > + }
> >
> > - if (evdet_irq > 0 && evdet_irq != client->irq)
> > + if (evdet_irq > 0 && evdet_irq != client->irq) {
> > ffffffff81aef656: 85 db test %ebx,%ebx
> > ffffffff81aef658: 7e 0d jle ffffffff81aef667 <isl1208_probe+0x1f0>
> > ffffffff81aef65a: 41 39 9d bc 01 00 00 cmp %ebx,0x1bc(%r13)
> > @@ -1663,6 +1664,7 @@ ffffffff81aef661: 0f 85 0a 01 00 00
> > rc = isl1208_setup_irq(client, evdet_irq);
> > if (rc)
> > return rc;
> > + }
> >
> > As you can see, no change in assembly but it is worse to read. gcc on
> > arm behaves the same way.
>
> Hi Alexandre,
> I am not sure that I fully understand your assembly code analysis. Maybe my patch comment was misleading, because I am not talking about a redundant test inside the if block, but ouside of it (after it).
>

I understood that and what I'm showing is that it doesn't matter to the
compiler, it will not test the same variable twice if it didn't change.

> Here is the original code with my annotations. Let's assume that the variable client->irq = 0:
>
> ---
> /* If client->irq = 0, then function isl1208_setup_irq() will not be called, and rc will not be modified: */
> if (client->irq > 0)
> rc = isl1208_setup_irq(client, client->irq);
>
> /* If rc hasn't been modified, there is no need to re-test its value here: */
> if (rc)
> return rc;
> ---
>
> After the patch, this code section becomes:
>
> ---
> if (client->irq > 0) {
> rc = isl1208_setup_irq(client, client->irq);
> if (rc)
> return rc;
> }
> ---
>
> For me it is more logical and clearer like this. Moreover, you can see that at line 873 of the original driver, the same kind of mechanism is used:
>
> ---
> if (isl1208->config->has_timestamp) {
> rc = rtc_add_group(isl1208->rtc, &isl1219_rtc_sysfs_files);
> if (rc)
> return rc;
> }
> ---
>
> Regards,
> Hugo.
>
>
> > > Signed-off-by: Hugo Villeneuve <[email protected]>
> > > ---
> > > drivers/rtc/rtc-isl1208.c | 14 ++++++++------
> > > 1 file changed, 8 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/drivers/rtc/rtc-isl1208.c b/drivers/rtc/rtc-isl1208.c
> > > index 182dfa605515..c7f04df5a0b6 100644
> > > --- a/drivers/rtc/rtc-isl1208.c
> > > +++ b/drivers/rtc/rtc-isl1208.c
> > > @@ -880,15 +880,17 @@ isl1208_probe(struct i2c_client *client, const struct i2c_device_id *id)
> > > if (rc)
> > > return rc;
> > >
> > > - if (client->irq > 0)
> > > + if (client->irq > 0) {
> > > rc = isl1208_setup_irq(client, client->irq);
> > > - if (rc)
> > > - return rc;
> > > + if (rc)
> > > + return rc;
> > > + }
> > >
> > > - if (evdet_irq > 0 && evdet_irq != client->irq)
> > > + if (evdet_irq > 0 && evdet_irq != client->irq) {
> > > rc = isl1208_setup_irq(client, evdet_irq);
> > > - if (rc)
> > > - return rc;
> > > + if (rc)
> > > + return rc;
> > > + }
> > >
> > > rc = devm_rtc_nvmem_register(isl1208->rtc, &isl1208->nvmem_config);
> > > if (rc)
> > > --
> > > 2.30.2
> > >
> >
> > --
> > Alexandre Belloni, co-owner and COO, Bootlin
> > Embedded Linux and Kernel engineering
> > https://bootlin.com
> >
>
>
> --
> Hugo Villeneuve <[email protected]>

--
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com