Push TxRate patches upstream

[silverjam]

Issue by silverjam from Monday Nov 04, 2013 at 18:36 GMT
Originally opened as https://github.com/cozybit/wcn36xx/issues/2

---

I think we can probably get the TxRate patches upstream. CY: Let me know if you think this is possible.

[silverjam]

Comment by chunyeow from Tuesday Nov 05, 2013 at 06:02 GMT

---

Sure, but I try to fix one by one first. Need Eugene to accept the mesh patches including the MCs rate first.

[silverjam]

Comment by silverjam from Tuesday Nov 05, 2013 at 06:06 GMT

---

OK.
On Nov 4, 2013 10:02 PM, "Chun-Yeow" [email protected] wrote:

Sure, but I try to fix one by one first. Need Eugene to accept the mesh
patches including the MCs rate first.

—
Reply to this email directly or view it on GitHubhttps://github.com/cozybit/wcn36xx/issues/2#issuecomment-27749631
.

[silverjam]

Comment by chunyeow from Friday Nov 08, 2013 at 01:06 GMT

---

Bob's patch is not upstream. So, the TxRate reporting makes no sense even if we upstream this. So is Bob going to upstream the patch?

[silverjam]

Comment by silverjam from Friday Nov 08, 2013 at 05:03 GMT

---

I'll ask, maybe we can push just that upstream, or push it as part of this.
On Nov 7, 2013 5:06 PM, "Chun-Yeow" [email protected] wrote:

Bob's patch is not upstream. So, the TxRate reporting makes no sense even
if we upstream this. So is Bob going to upstream the patch?

—
Reply to this email directly or view it on GitHubhttps://github.com/cozybit/wcn36xx/issues/2#issuecomment-28022839
.

[silverjam]

Comment by twpedersen from Friday Nov 08, 2013 at 05:10 GMT

---

You'll have to push both as part of the same set since I don't think
they'll take a new driver API without any users.

On Thu, Nov 7, 2013 at 9:03 PM, Jason Mobarak [email protected]:

I'll ask, maybe we can push just that upstream, or push it as part of
this.
On Nov 7, 2013 5:06 PM, "Chun-Yeow" [email protected] wrote:

Bob's patch is not upstream. So, the TxRate reporting makes no sense
even
if we upstream this. So is Bob going to upstream the patch?

—
Reply to this email directly or view it on GitHub<
https://github.com/cozybit/wcn36xx/issues/2#issuecomment-28022839>
.

—
Reply to this email directly or view it on GitHubhttps://github.com/cozybit/wcn36xx/issues/2#issuecomment-28036216
.

Thomas

[silverjam]

Comment by bcopeland from Friday Nov 08, 2013 at 14:21 GMT

---

On Thu, Nov 07, 2013 at 09:10:06PM -0800, twpedersen wrote:

You'll have to push both as part of the same set since I don't think
they'll take a new driver API without any users.

Yes. Also, I now can't remember why my patch passed only the
address instead of the sta. That should probably be fixed.

[silverjam]

Comment by twpedersen from Friday Nov 08, 2013 at 14:33 GMT

---

On Nov 8, 2013 6:22 AM, "Bob Copeland" [email protected] wrote:

On Thu, Nov 07, 2013 at 09:10:06PM -0800, twpedersen wrote:

You'll have to push both as part of the same set since I don't think
they'll take a new driver API without any users.

Yes. Also, I now can't remember why my patch passed only the
address instead of the sta. That should probably be fixed.

Because SDIO needs to sleep, but the sta needs to be under rcu lock.

[silverjam]

Comment by bcopeland from Friday Nov 08, 2013 at 14:41 GMT

---

On Fri, Nov 08, 2013 at 06:33:03AM -0800, twpedersen wrote:

Yes. Also, I now can't remember why my patch passed only the
address instead of the sta. That should probably be fixed.

Because SDIO needs to sleep, but the sta needs to be under rcu lock.

Oh, yeah. :(

Bob Copeland %% www.bobcopeland.com