This is a summary of discussions relative to the merge request created by Eric 
Gallager (egallager) titled
Add news announcement of AI Policy
since its creation.

Description: There is currently no link to the [AI 
Policy](https://gcc.gnu.org/ai-policy.html) from the main page. This improves 
the situation a bit by adding a news entry for it.

The full and up to date discussion can be found at 
https://forge.sourceware.org/gcc/gcc-wwwdocs/pulls/3

The merge request has been closed without being merged directly on the forge 
repository.


On 2026-09-10 20:40:18+00:00, Eric Gallager (egallager) wrote:

oops wait that's more commits than I meant to include 


On 2026-09-10 22:36:40+00:00, Eric Gallager (egallager) wrote:

@forge-bot wrote in 
https://forge.sourceware.org/gcc/gcc-wwwdocs/pulls/3#issuecomment-7441:

> Sent patch series version 1 containing 43 patches to @egallager's email 
> address. Message ID is 
> `<bmm.hm6a02tws8.gcc.gcc-wwwdocs.egallager.3....@forge-stage.sourceware.org>`.

Aaaaaaa that was too many; some of them ended up getting tagged as spam... 


On 2026-09-11 08:13:33+00:00, Jonathan Wakely (redi) wrote:

I've synced the gcc-wwwdocs forge repo with upstream so there's only one commit 
here now

On 2026-09-11 16:40:39+00:00, Jonathan Wakely (redi) <[email protected]> 
commented on the code:


> +++ htdocs/index.html
> @@ -56,6 +56,11 @@ mission statement</a>.</p>
>  <h2 id="news">News</h2>
>  <dl>
>  
> +<dt><span>The <a href="ai-policy.html"></a>GCC AI Poilcy</a>
"Poilcy" typo
> +++ htdocs/index.html
> @@ -56,6 +56,11 @@ mission statement</a>.</p>
>  <h2 id="news">News</h2>
>  <dl>
>  
> +<dt><span>The <a href="ai-policy.html"></a>GCC AI Poilcy</a>
Fixed in 23ff006b25


On 2026-09-11 16:41:25+00:00, Jonathan Wakely (redi) <[email protected]> 
commented on the code:


> +++ htdocs/index.html
> @@ -57,2 +57,4 @@
>  <dl>
>  
> +<dt><span>The <a href="ai-policy.html"></a>GCC AI Poilcy</a>
> +  <a href="https://gcc.gnu.org/pipermail/gcc/2026-July/248628.html";>is 
> announced.</a></span>
I think we want past tense here, "was announced". Or just "announced".
> +++ htdocs/index.html
> @@ -57,2 +57,4 @@
>  <dl>
>  
> +<dt><span>The <a href="ai-policy.html"></a>GCC AI Poilcy</a>
> +  <a href="https://gcc.gnu.org/pipermail/gcc/2026-July/248628.html";>is 
> announced.</a></span>
I went with the latter in 23ff006b25.

On 2026-09-11 20:14:32+00:00, Jonathan Wakely (redi) <[email protected]> 
approved the changes:



On 2026-09-11 20:15:07+00:00, Jonathan Wakely (redi) <[email protected]> 
requested changes to the code:


> +++ htdocs/index.html
> @@ -58,1 +58,4 @@
>  
> +<dt><span>The <a href="ai-policy.html"></a>GCC AI Policy</a>
> +  <a 
> href="https://gcc.gnu.org/pipermail/gcc/2026-July/248628.html";>announced.</a></span>
> +    <span class="date">[2026-29-07]</span></dt>
what kind of cursed date is this?!
> +++ htdocs/index.html
> @@ -58,1 +58,4 @@
>  
> +<dt><span>The <a href="ai-policy.html"></a>GCC AI Policy</a>
> +  <a 
> href="https://gcc.gnu.org/pipermail/gcc/2026-July/248628.html";>announced.</a></span>
> +    <span class="date">[2026-29-07]</span></dt>
The result of a momentary brain failure. Fixed in 8a166899b6.

On 2026-09-16 23:22:36+00:00, Eric Gallager (egallager) requested that Jonathan 
Wakely (redi) <[email protected]> review the code:





On 2026-09-18 14:27:58+00:00, Jonathan Wakely (redi) wrote:

Please squash the three commits into one and then this is OK to push to wwwdocs.


On 2026-09-20 00:55:33+00:00, Eric Gallager (egallager) wrote:

@redi wrote in 
https://forge.sourceware.org/gcc/gcc-wwwdocs/pulls/3#issuecomment-7522:

> Please squash the three commits into one and then this is OK to push to 
> wwwdocs.

Is there a way to do that via the Forge interface? I know GitHub has an option 
to squash pull requests when merging them, and that Forgejo generally attempts 
to be compatible with GitHub, but since I'm not authorized to merge this pull 
request, I'm not able to see if that option is also available here... 


On 2026-09-20 08:48:39+00:00, Jonathan Wakely (redi) wrote:

Nobody should be merging pull requests here anyway, the master branch here is a 
read only mirror of the real repo. Changes need to be committed to the real 
repo, not here. 

So squash the commits on the command line then push to the real repo from the 
command line. 

https://gcc.gnu.org/wiki/ForgeExperiment#Monitoring_and_merging_changes


On 2026-09-20 12:34:08+00:00, (yuao) wrote:

@redi wrote in 
https://forge.sourceware.org/gcc/gcc-wwwdocs/pulls/3#issuecomment-7533:

> Nobody should be merging pull requests here anyway, the master branch here is 
> a read only mirror of the real repo. Changes need to be committed to the real 
> repo, not here.
> 
> So squash the commits on the command line then push to the real repo from the 
> command line.
> 
> https://gcc.gnu.org/wiki/ForgeExperiment#Monitoring_and_merging_changes

I'm wondering if we could do something similar to what openjdk does, which has 
a bot that help you do a lot of task, including integrate a PR. Like this one: 
https://github.com/openjdk/jdk/pull/32721#issuecomment-5749064779

And the bot itself is open-sourced: https://github.com/openjdk/skara


On 2026-09-21 13:11:22+00:00, Jonathan Wakely (redi) wrote:

Such suggestions should be discussed on the mailing lists not in comments on a 
random PR


On 2026-09-21 16:48:30+00:00, (yuao) wrote:

@redi wrote in 
https://forge.sourceware.org/gcc/gcc-wwwdocs/pulls/3#issuecomment-7538:

> Such suggestions should be discussed on the mailing lists not in comments on 
> a random PR

Sorry for the inappropriate location. Still I'm not intended to comment in a 
random PR but this looks like a concrete example of how we can benefit from the 
proposed feature.

BTW, which mailing list is more suitable for discuss this? `forge` or `gcc`?


On 2026-09-21 18:15:52+00:00, Jonathan Wakely (redi) wrote:

Maybe the forge list initially, but eventually there will need to be wider 
agreement that it's ok to automate merging commits from this repo to the real 
one. 


On 2026-09-22 15:37:18+00:00, Jonathan Wakely (redi) wrote:

The html wasn't valid (too many `</a>` tags). Fixed and pushed as 
e986d2ae463ec95b51ca4920cbd71dc034f3c7b0

Reply via email to