Skip to contentSkip to conversation

staging: wilc1000: Use crc7 in lib/ rather than a private copy

PATCH In review [email protected] 4 participants 9 messages Mar 26, 2020 – Apr 4, 2020 Staging

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 staging: wilc1000: Use crc7 in lib/ rather than a private copy

From: George Spelvin <[email protected]>

The code in lib/ is the desired polynomial, and even includes the 1-bit left shift in the table rather than needing to code it explicitly.

While I'm in Kconfig, add a description of what a WILC1000 is. Kconfig questions that require me to look up a data sheet to find out that I probably don't have one are a pet peeve.

---
v3: added 'staging' to subject and submit to [email protected]

v2: Rebase on the staging-next tree

 drivers/staging/wilc1000/Kconfig |  5 +++
 drivers/staging/wilc1000/spi.c   | 64 +++-----------------------------
 2 files changed, 11 insertions(+), 58 deletions(-)
drivers/staging/wilc1000/Kconfig +5 -0
index 59e58550d139..80c92e8bf8a5 100644
@@ -2,6 +2,10 @@
 config WILC1000
 	tristate
 	help
+	  Add support for the Atmel WILC1000 802.11 b/g/n SoC.
+	  This provides Wi-FI over an SDIO or SPI interface, and
+	  is usually found in IoT devices.
+
 	  This module only support IEEE 802.11n WiFi.
 
 config WILC1000_SDIO
@@ -22,6 +26,7 @@ config WILC1000_SPI
 	tristate "Atmel WILC1000 SPI (WiFi only)"
 	depends on CFG80211 && INET && SPI
 	select WILC1000
+	select CRC7
 	help
 	  This module adds support for the SPI interface of adapters using
 	  WILC1000 chipset. The Atmel WILC1000 has a Serial Peripheral
drivers/staging/wilc1000/spi.c +6 -58

82 lines not on this page · raw message

Signature · 1 line
2.24.0
#1 lore raw
Quoted 10 lines · On Thu, Mar 26, 2020 at 03:23:36PM +0000, [email protected] wrote:

From: George Spelvin <[email protected]>

The code in lib/ is the desired polynomial, and even includes the 1-bit left shift in the table rather than needing to code it explicitly.

While I'm in Kconfig, add a description of what a WILC1000 is. Kconfig questions that require me to look up a data sheet to find out that I probably don't have one are a pet peeve.

I don't know how this patch made it through two versions without anyone complaining that this paragraph should be done as a separate patch...

Quoted 5 lines

---

This should have you Signed-off-by. The Reviewed-by is kind of assumed so you can drop that bit. But everyone who touches a patch needs to add their signed off by.

regards,
dan carpenter

#2 lore raw

Hi Dan,

Quoted 9 lines · Dan Carpenter

EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe

Quoted 6 lines · On Thu, Mar 26, 2020 at 03:23:36PM +0000, [email protected] wrote:

From: George Spelvin <[email protected]>

The code in lib/ is the desired polynomial, and even includes the 1-bit left shift in the table rather than needing to code it explicitly.

Quoted 8 lines
Quoted 4 lines

While I'm in Kconfig, add a description of what a WILC1000 is. Kconfig questions that require me to look up a data sheet to find out that I probably don't have one are a pet peeve.

I don't know how this patch made it through two versions without anyone complaining that this paragraph should be done as a separate patch...

The first two version of patches were not submitted to devel@driverdev mailing list. I too missed the point to keep the Kconfig changes in separate patch.

Quoted 10 lines
Quoted 5 lines

---

This should have you Signed-off-by. The Reviewed-by is kind of assumed so you can drop that bit. But everyone who touches a patch needs to add their signed off by.

Thanks, I will make a note of it.

Regards
Ajay

#3 lore raw
In reply to Dan Carpenter
Quoted 2 lines · Dan Carpenter

I don't know how this patch made it through two versions without anyone complaining that this paragraph should be done as a separate patch...

I often fold comment (and spacing/formatting) patches in to a main patch, when touching adjacent code anyway and it doesn't cause distracting clutter.

This seemed like such a case, which is why I submitted it as one. But it's a bit of style thing.

Quoted 9 lines
Quoted 5 lines

---

This should have you Signed-off-by. The Reviewed-by is kind of assumed so you can drop that bit. But everyone who touches a patch needs to add their signed off by.

Er... all he did was add "staging: " to the front of the title.

That's not a change to the code at all, and as trivial a change to the commit message as adding "Reviewed-by:" to the end. We don't need S-o-b for such things or we'd end up in a horrible infinite recursion.

#4 lore raw
Quoted 11 lines · George Spelvin
Quoted 2 lines · Dan Carpenter

I don't know how this patch made it through two versions without anyone complaining that this paragraph should be done as a separate patch...

I often fold comment (and spacing/formatting) patches in to a main patch, when touching adjacent code anyway and it doesn't cause distracting clutter.

This seemed like such a case, which is why I submitted it as one. But it's a bit of style thing.

We're super strict in Staging. :P Greg is more strict than I am.

Quoted 10 lines
Quoted 9 lines
Quoted 5 lines

---

This should have you Signed-off-by. The Reviewed-by is kind of assumed so you can drop that bit. But everyone who touches a patch needs to add their signed off by.

Quoted 6 lines

Er... all he did was add "staging: " to the front of the title.

That's not a change to the code at all, and as trivial a change to the commit message as adding "Reviewed-by:" to the end. We don't need S-o-b for such things or we'd end up in a horrible infinite recursion.

You've misunderstood. He sent the email so he has to add his Signed-off-by. It's not at all related to changing anything in the patch. That's how sign offs work.

regards,
dan carpenter

#5 lore raw
Quoted 9 lines · Dan Carpenter
Quoted 8 lines · George Spelvin
Quoted 2 lines · Dan Carpenter

I don't know how this patch made it through two versions without anyone complaining that this paragraph should be done as a separate patch...

I often fold comment (and spacing/formatting) patches in to a main patch, when touching adjacent code anyway and it doesn't cause distracting clutter.

Quoted 5 lines
Quoted 3 lines

This seemed like such a case, which is why I submitted it as one. But it's a bit of style thing.

We're super strict in Staging. :P Greg is more strict than I am.

Okay, but it's my fault, not his.

Quoted 6 lines
Quoted 6 lines

This should have you Signed-off-by. The Reviewed-by is kind of assumed

Quoted 2 lines

so you can drop that bit. But everyone who touches a patch needs to add their signed off by.

Er... all he did was add "staging: " to the front of the title.

Quoted 8 lines
Quoted 4 lines

That's not a change to the code at all, and as trivial a change to the commit message as adding "Reviewed-by:" to the end. We don't need S-o-b for such things or we'd end up in a horrible infinite recursion.

You've misunderstood. He sent the email so he has to add his Signed-off-by. It's not at all related to changing anything in the patch. That's how sign offs work.

Looking at my commits (just because I remember how they went in), you seem to be right, but damn, submitting-patches.rst could be clearer on the subject.

I understand that it's addressed more to patch authors than maintainers forwarding them, but I've read that thing a dozen times, and the description of S-o-b always seemed to be about copyright.

So I had assumed that edits which were below the de minimus standard of copyright didn't need a separate S-o-b.

Am I right that there should be an S-o-b from everyone from the patch author to the patch committer (as recorded in git)? And the one exception is that we don't need S-o-b for git pulls after that, because the merge commits record the information?

For example, my patch series ending at 4684fe95300c (v4.7-rc1~8^2) only has my S-o-b because it was pulled straight from my git server and merge 7e0fb73c52c4 (v4.7-rc1~8) records who merged it.

But b5c56e0cdd62 has an S-o-b from both akpm and Linus because it went to akpm, into his quilt, and then as a patch series to Linus, who committed it.

All of which is eactly why git-am has a -s option.

That's not a hard rule to understand, but I wish submitting-patches *said* so somewhere, rather than having it be implied by the existence of option (c) in the DCO and the fact that it's *doesn't* say that someone else's S-o-b will suffice.

And the git merge exception should be stated, because otherwise it's not clear what the limits of that exception are. I had assumed that accumulating and forwarding patches in general was okay without a S-o-b.

So thank you for enlightening me, and if you can confirm the rules, I'll prepare a Documentation/ patch to reduce re-occurrence.

#6 lore raw
Quoted 41 lines · George Spelvin
Quoted 9 lines · Dan Carpenter
Quoted 8 lines · George Spelvin
Quoted 2 lines · Dan Carpenter

I don't know how this patch made it through two versions without anyone complaining that this paragraph should be done as a separate patch...

I often fold comment (and spacing/formatting) patches in to a main patch, when touching adjacent code anyway and it doesn't cause distracting clutter.

Quoted 5 lines
Quoted 3 lines

This seemed like such a case, which is why I submitted it as one. But it's a bit of style thing.

We're super strict in Staging. :P Greg is more strict than I am.

Okay, but it's my fault, not his.

Quoted 6 lines
Quoted 6 lines

This should have you Signed-off-by. The Reviewed-by is kind of assumed

Quoted 2 lines

so you can drop that bit. But everyone who touches a patch needs to add their signed off by.

Er... all he did was add "staging: " to the front of the title.

Quoted 8 lines
Quoted 4 lines

That's not a change to the code at all, and as trivial a change to the commit message as adding "Reviewed-by:" to the end. We don't need S-o-b for such things or we'd end up in a horrible infinite recursion.

You've misunderstood. He sent the email so he has to add his Signed-off-by. It's not at all related to changing anything in the patch. That's how sign offs work.

Looking at my commits (just because I remember how they went in), you seem to be right, but damn, submitting-patches.rst could be clearer on the subject.

I understand that it's addressed more to patch authors than maintainers forwarding them, but I've read that thing a dozen times, and the description of S-o-b always seemed to be about copyright.

Quoted 5 lines

So I had assumed that edits which were below the de minimus standard of copyright didn't need a separate S-o-b.

Am I right that there should be an S-o-b from everyone from the patch author to the patch committer (as recorded in git)?

Yes, everyone either modifying or distributing (=submitting) the patch should add their s-o-b.

Quoted 2 lines

And the one exception is that we don't need S-o-b for git pulls after that, because the merge commits record the information?

Correct. When you do a git pull you are pulling the commits without any modification, so technically it's not even possible to add the s-o-b lines to the commits you are pulling. Modifying the commit logs would need a rebase and then it not would be a normal git pull anymore.

Signature · 1 line
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
#7 lore raw
In reply to George Spelvin
Quoted 4 lines · George Spelvin

I understand that it's addressed more to patch authors than maintainers forwarding them, but I've read that thing a dozen times, and the description of S-o-b always seemed to be about copyright.

It's to say that you didn't add anything which you shouldn't have, for example, secret SCO UnixWare stuff.

Quoted 7 lines

So I had assumed that edits which were below the de minimus standard of copyright didn't need a separate S-o-b.

Am I right that there should be an S-o-b from everyone from the patch author to the patch committer (as recorded in git)? And the one exception is that we don't need S-o-b for git pulls after that, because the merge commits record the information?

Yes. Also if people added their S-o-b for git merges it would change the git hash for the patch which would suck.

regards,
dan carpenter

#8 lore raw
Quoted 7 lines · Dan Carpenter
Quoted 3 lines · George Spelvin

I understand that it's addressed more to patch authors than maintainers forwarding them, but I've read that thing a dozen times, and the description of S-o-b always seemed to be about copyright.

It's to say that you didn't add anything which you shouldn't have, for example, secret SCO UnixWare stuff.

Yes, I'm familiar with the (irritating) history. Which is why I had the idea stuck in my head that that it was all about copyright and if you didn't add anything copyrightable, an S-o-b wasn't required. No more than I'd ask for one from the administrator of the e-mail system which delivered it.

submitting-patches.rst says "sign your work". It didn't occur to me to sign something that wasn't my work.

Quoted 10 lines
Quoted 7 lines

So I had assumed that edits which were below the de minimus standard of copyright didn't need a separate S-o-b.

Am I right that there should be an S-o-b from everyone from the patch author to the patch committer (as recorded in git)? And the one exception is that we don't need S-o-b for git pulls after that, because the merge commits record the information?

Yes. Also if people added their S-o-b for git merges it would change the git hash for the patch which would suck.

I understand the technical difficulties, but lawyers aren't always deterred by such things. :-) Seriously, it's clear there has to be an exception; the question was about the scope of the exception.

Thank you for your patience clarifying this stuff for the nth time.