/
gentoo-kernel.2021-07-25-13.01.log.html
287 lines (282 loc) · 38.1 KB
/
gentoo-kernel.2021-07-25-13.01.log.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>#gentoo-kernel log</title>
<style type="text/css">
/* For the .log.html */
pre { /*line-height: 125%;*/
white-space: pre-wrap; }
body { background: #f0f0f0; }
body .tm { color: #007020 } /* time */
body .nk { color: #062873; font-weight: bold } /* nick, regular */
body .nka { color: #007020; font-weight: bold } /* action nick */
body .ac { color: #00A000 } /* action line */
body .hi { color: #4070a0 } /* hilights */
/* Things to make particular MeetBot commands stick out */
body .topic { color: #007020; font-weight: bold }
body .topicline { color: #000080; font-weight: bold }
body .cmd { color: #007020; font-weight: bold }
body .cmdline { font-weight: bold }
a {text-decoration:none;}
a:hover {text-decoration:underline;}
</style>
</head>
<body>
<pre><a href="#l-1" name="l-1"><span class="tm">13:01</span></a><span class="nk"> <alicef></span> <span class="cmd">#startmeeting</span><span class="cmdline"></span>
<a href="#l-2" name="l-2"><span class="tm">13:01</span></a><span class="nk"> <AlicefBot></span> Meeting started at 13:01:41 UTC. The chair is alicef. Information about MeetBot at https://github.com/GKernelCI/Gmeetbot
<a href="#l-3" name="l-3"><span class="tm">13:01</span></a><span class="nk"> <AlicefBot></span> Available commands: action, commands, idea, info, link, nick
<a href="#l-4" name="l-4"><span class="tm">13:02</span></a><span class="nk"> <alicef></span> rollcall
<a href="#l-5" name="l-5"><span class="tm">13:02</span></a><span class="nk"> <alicef></span> !team kernel
<a href="#l-6" name="l-6"><span class="tm">13:02</span></a><span class="nk"> <alicef></span> <span class="hi">mpagano:</span> kveremitz Whissi
<a href="#l-7" name="l-7"><span class="tm">13:03</span></a><span class="nk"> <alicef></span> <span class="topic">#topic </span><span class="topicline">rollcall</span>
<a href="#l-8" name="l-8"><span class="tm">13:03</span></a><span class="nk"> <alicef></span> please say hi if you are around
<a href="#l-9" name="l-9"><span class="tm">13:06</span></a><span class="nk"> <alicef></span> hi
<a href="#l-10" name="l-10"><span class="tm">13:07</span></a><span class="nk"> <alicef></span> anyone around ?
<a href="#l-11" name="l-11"><span class="tm">13:08</span></a><span class="nk"> <mgorny></span> i'm around ;-P
<a name="l-12"></a><span class="tm">13:08 </span><span class="nka">* mgorny</span> <span class="ac">hides</span>
<a href="#l-13" name="l-13"><span class="tm">13:09</span></a><span class="nk"> <alicef></span> :)
<a href="#l-14" name="l-14"><span class="tm">13:09</span></a><span class="nk"> <alicef></span> is dinner time there?
<a href="#l-15" name="l-15"><span class="tm">13:10</span></a><span class="nk"> <montjoie></span> hi
<a href="#l-16" name="l-16"><span class="tm">13:10</span></a><span class="nk"> <mgorny></span> in EU? it's 3 PM, so some people eat around the time, yeah
<a href="#l-17" name="l-17"><span class="tm">13:10</span></a><span class="nk"> <alicef></span> hi montjoie :)
<a href="#l-18" name="l-18"><span class="tm">13:10</span></a><span class="nk"> <alicef></span> !time mgorny
<a href="#l-19" name="l-19"><span class="tm">13:10</span></a><span class="nk"> <willikins></span> <span class="hi">alicef:</span> Europe - Warsaw - Sun Jul 25 15:10 CEST
<a href="#l-20" name="l-20"><span class="tm">13:11</span></a><span class="nk"> <alicef></span> is night here after dinner time
<a href="#l-21" name="l-21"><span class="tm">13:11</span></a><span class="nk"> <alicef></span> but I didn't eat yet
<a href="#l-22" name="l-22"><span class="tm">13:11</span></a><span class="nk"> <alicef></span> <span class="hi">mpagano:</span> kveremitz Whissi
<a href="#l-23" name="l-23"><span class="tm">13:12</span></a><span class="nk"> <alicef></span> !proj kernel
<a href="#l-24" name="l-24"><span class="tm">13:12</span></a><span class="nk"> <willikins></span> (kernel@gentoo.org) alicef, blueness, chainsaw, mpagano, whissi, zorry
<a href="#l-25" name="l-25"><span class="tm">13:12</span></a><span class="nk"> <alicef></span> oooh it worked
<a href="#l-26" name="l-26"><span class="tm">13:12</span></a><span class="nk"> <alicef></span> mgorny everything ok ?
<a href="#l-27" name="l-27"><span class="tm">13:13</span></a><span class="nk"> <mgorny></span> i suppose so
<a href="#l-28" name="l-28"><span class="tm">13:13</span></a><span class="nk"> <mgorny></span> you?
<a href="#l-29" name="l-29"><span class="tm">13:13</span></a><span class="nk"> <alicef></span> not bad :)
<a href="#l-30" name="l-30"><span class="tm">13:13</span></a><span class="nk"> <AlicefBot></span> News from kernel: 5.13.5: stable <http://www.kernel.org/>
<a href="#l-31" name="l-31"><span class="tm">13:13</span></a><span class="nk"> <AlicefBot></span> News from kernel: 5.10.53: longterm <http://www.kernel.org/>
<a href="#l-32" name="l-32"><span class="tm">13:13</span></a><span class="nk"> <AlicefBot></span> News from kernel: 5.4.135: longterm <http://www.kernel.org/>
<a href="#l-33" name="l-33"><span class="tm">13:13</span></a><span class="nk"> <alicef></span> eh ? as we do the meeting ...
<a href="#l-34" name="l-34"><span class="tm">13:14</span></a><span class="nk"> <alicef></span> I think we should wait at least mpagano
<a href="#l-35" name="l-35"><span class="tm">13:14</span></a><span class="nk"> <alicef></span> for the meeting to start
<a href="#l-36" name="l-36"><span class="tm">13:15</span></a><span class="nk"> <mgorny></span> <span class="hi">alicef:</span> eat something while you wait ;-)
<a name="l-37"></a><span class="tm">13:16 </span><span class="nka">* alicef</span> <span class="ac">is opening a peanuts can</span>
<a href="#l-38" name="l-38"><span class="tm">13:16</span></a><span class="nk"> <alicef></span> I think we can at least start with gkernelci
<a href="#l-39" name="l-39"><span class="tm">13:16</span></a><span class="nk"> <alicef></span> <span class="topic">#topic </span><span class="topicline">gkernelci</span>
<a href="#l-40" name="l-40"><span class="tm">13:17</span></a><span class="nk"> <alicef></span> not so much happened. I did some small code fixes and updated the kernel version
<a href="#l-41" name="l-41"><span class="tm">13:17</span></a><span class="nk"> <alicef></span> now is supporting 5.13 and next 5.14
<a href="#l-42" name="l-42"><span class="tm">13:18</span></a><span class="nk"> <alicef></span> and I requested a dedicated server as the compilation time is becoming a bootle neck
<a href="#l-43" name="l-43"><span class="tm">13:18</span></a><span class="nk"> <alicef></span> mgorny did you give a look a gkernelci ? do you have any opinion ?
<a href="#l-44" name="l-44"><span class="tm">13:19</span></a><span class="nk"> <mgorny></span> only a brief look at some point
<a href="#l-45" name="l-45"><span class="tm">13:19</span></a><span class="nk"> <mgorny></span> inspired me for src_test() in dist-kernel
<a href="#l-46" name="l-46"><span class="tm">13:19</span></a><span class="nk"> <alicef></span> is useless ? :/
<a href="#l-47" name="l-47"><span class="tm">13:19</span></a><span class="nk"> <alicef></span> oh nice :)
<a href="#l-48" name="l-48"><span class="tm">13:19</span></a><span class="nk"> <mgorny></span> at some point would be nice to deduplicate work
<a href="#l-49" name="l-49"><span class="tm">13:19</span></a><span class="nk"> <mgorny></span> since i need to build+test 5.4/5.10/5.13 anyway for gentoo-kernel-bin
<a href="#l-50" name="l-50"><span class="tm">13:20</span></a><span class="nk"> <mgorny></span> takes around 20 min per kernel version per arch on my pc
<a href="#l-51" name="l-51"><span class="tm">13:20</span></a><span class="nk"> <alicef></span> are you doing also kselftest ?
<a href="#l-52" name="l-52"><span class="tm">13:22</span></a><span class="nk"> <alicef></span> are you thinking that by building and testing gentoo-kernel-bin gkernelci become useless ?
<a href="#l-53" name="l-53"><span class="tm">13:22</span></a><span class="nk"> <alicef></span> gentoo-sources need anyway to be tested before the release
<a href="#l-54" name="l-54"><span class="tm">13:22</span></a><span class="nk"> <alicef></span> and that is what is currently doing gkernelci
<a href="#l-55" name="l-55"><span class="tm">13:23</span></a><span class="nk"> <mgorny></span> no, not yet
<a href="#l-56" name="l-56"><span class="tm">13:23</span></a><span class="nk"> <alicef></span> that not yet is scary :P
<a href="#l-57" name="l-57"><span class="tm">13:23</span></a><span class="nk"> <mgorny></span> i need to investigate kselftest at some point
<a href="#l-58" name="l-58"><span class="tm">13:23</span></a><span class="nk"> <mgorny></span> (i was replying wrt kselftest)
<a href="#l-59" name="l-59"><span class="tm">13:24</span></a><span class="nk"> <alicef></span> ah is about kselftest :)
<a href="#l-60" name="l-60"><span class="tm">13:24</span></a><span class="nk"> <alicef></span> what do you mean by deduplicate work ?
<a href="#l-61" name="l-61"><span class="tm">13:25</span></a><span class="nk"> <mgorny></span> not sure yet
<a href="#l-62" name="l-62"><span class="tm">13:25</span></a><span class="nk"> <mgorny></span> would be nice not to do the same thing twice
<a href="#l-63" name="l-63"><span class="tm">13:25</span></a><span class="nk"> <alicef></span> yes but we are still releasing two packages
<a href="#l-64" name="l-64"><span class="tm">13:25</span></a><span class="nk"> <mgorny></span> either make gkernelci build gentoo-kernel images, or improve src_test() to make it equiv to gkernelci
<a href="#l-65" name="l-65"><span class="tm">13:26</span></a><span class="nk"> <alicef></span> also if you improve src_test() we need still to test each linux-patches commit
<a href="#l-66" name="l-66"><span class="tm">13:26</span></a><span class="nk"> <mgorny></span> how does gkernelci test new kernel versions? do you start it before pushing gentoo-sources or after?
<a href="#l-67" name="l-67"><span class="tm">13:26</span></a><span class="nk"> <alicef></span> is doing both
<a href="#l-68" name="l-68"><span class="tm">13:27</span></a><span class="nk"> <mgorny></span> ah, ok
<a href="#l-69" name="l-69"><span class="tm">13:27</span></a><span class="nk"> <alicef></span> but the test of pushing gentoo-sources is still not yet finished
<a href="#l-70" name="l-70"><span class="tm">13:27</span></a><span class="nk"> <mgorny></span> i suppose no good solution then yet
<a href="#l-71" name="l-71"><span class="tm">13:27</span></a><span class="nk"> <alicef></span> is actually testing everything under sys-kernel/*
<a href="#l-72" name="l-72"><span class="tm">13:28</span></a><span class="nk"> <alicef></span> but currently is just using ebuild
<a href="#l-73" name="l-73"><span class="tm">13:28</span></a><span class="nk"> <mgorny></span> we would find it helpful to have access to new releases of genpatches early though
<a href="#l-74" name="l-74"><span class="tm">13:28</span></a><span class="nk"> <mgorny></span> so we could start testing and building gentoo-kernel before gentoo-sources are pushed
<a href="#l-75" name="l-75"><span class="tm">13:29</span></a><span class="nk"> <alicef></span> when gentpatches are released we are making also gentoo-sources
<a href="#l-76" name="l-76"><span class="tm">13:29</span></a><span class="nk"> <alicef></span> usually at same time
<a href="#l-77" name="l-77"><span class="tm">13:29</span></a><span class="nk"> <mgorny></span> ah, ok
<a href="#l-78" name="l-78"><span class="tm">13:30</span></a><span class="nk"> <mgorny></span> i thought you were testing it first
<a href="#l-79" name="l-79"><span class="tm">13:30</span></a><span class="nk"> <alicef></span> gkernelci is testing each commit on linux-patches
<a href="#l-80" name="l-80"><span class="tm">13:30</span></a><span class="nk"> <mgorny></span> ah
<a href="#l-81" name="l-81"><span class="tm">13:30</span></a><span class="nk"> <alicef></span> linux-patches get build and tested
<a href="#l-82" name="l-82"><span class="tm">13:30</span></a><span class="nk"> <alicef></span> if they work we make genpatches
<a href="#l-83" name="l-83"><span class="tm">13:30</span></a><span class="nk"> <alicef></span> than gentoo-sources
<a href="#l-84" name="l-84"><span class="tm">13:30</span></a><span class="nk"> <mgorny></span> is it doing every single commit or skipping commits if many happen at once?
<a href="#l-85" name="l-85"><span class="tm">13:31</span></a><span class="nk"> <alicef></span> becouse in some cases we broke linux-patches
<a href="#l-86" name="l-86"><span class="tm">13:31</span></a><span class="nk"> <alicef></span> like I see few mangled commit but it happened
<a href="#l-87" name="l-87"><span class="tm">13:31</span></a><span class="nk"> <alicef></span> and gkernelci can catch such things
<a href="#l-88" name="l-88"><span class="tm">13:32</span></a><span class="nk"> <alicef></span> I think is doing each commit
<a href="#l-89" name="l-89"><span class="tm">13:33</span></a><span class="nk"> <alicef></span> is rare to have many commits happen at once
<a href="#l-90" name="l-90"><span class="tm">13:33</span></a><span class="nk"> <mgorny></span> btw shouldn't the topic be updated for 5.12 EOL?
<a href="#l-91" name="l-91"><span class="tm">13:34</span></a><span class="nk"> <alicef></span> <span class="hi">mgorny:</span> you are right !
<a href="#l-92" name="l-92"><span class="tm">13:34</span></a><span class="nk"> <alicef></span> but I have no idea how to change a topic during a meeting. bit worried that it mangle the bot
<a href="#l-93" name="l-93"><span class="tm">13:34</span></a><span class="nk"> <alicef></span> :P
<a href="#l-94" name="l-94"><span class="tm">13:36</span></a><span class="nk"> <mgorny></span> there's one more suggestion from us
<a href="#l-95" name="l-95"><span class="tm">13:36</span></a><span class="nk"> <mgorny></span> we think it would be better if genpatches version carried the full kernel version
<a href="#l-96" name="l-96"><span class="tm">13:36</span></a><span class="nk"> <mgorny></span> i.e. instead of genpatches version 22 correspoinding to 5.12.19
<a href="#l-97" name="l-97"><span class="tm">13:36</span></a><span class="nk"> <mgorny></span> it would be version 19
<a href="#l-98" name="l-98"><span class="tm">13:36</span></a><span class="nk"> <mgorny></span> and then 19.1, 19.2 etc. if need be
<a href="#l-99" name="l-99"><span class="tm">13:37</span></a><span class="nk"> <mgorny></span> right now it's hard to guess which genpatches version corresponds to which kernel version
<a href="#l-100" name="l-100"><span class="tm">13:37</span></a><span class="nk"> <alicef></span> I think in this case deduplicate work is not a problem. maybe src_test could keep do something that can help the user testing the kernel? and GkernelCI can keep working on testing the kernel on the developer part of things
<a href="#l-101" name="l-101"><span class="tm">13:39</span></a><span class="nk"> <alicef></span> I'm actually not sure how gentoo-kernel-bin work
<a href="#l-102" name="l-102"><span class="tm">13:39</span></a><span class="nk"> <alicef></span> as I'm honestly not using it
<a href="#l-103" name="l-103"><span class="tm">13:40</span></a><span class="nk"> <mgorny></span> it installs binary kernel + modules and compiles a minimal source tree needed to build kernel modules
<a href="#l-104" name="l-104"><span class="tm">13:41</span></a><span class="nk"> <alicef></span> and is using genpatches?
<a href="#l-105" name="l-105"><span class="tm">13:41</span></a><span class="nk"> <mgorny></span> yes
<a href="#l-106" name="l-106"><span class="tm">13:42</span></a><span class="nk"> <alicef></span> GENPATCHES_P=genpatches-${PV%.*}-$(( ${PV##*.} + 2 ))
<a href="#l-107" name="l-107"><span class="tm">13:42</span></a><span class="nk"> <alicef></span> I see
<a href="#l-108" name="l-108"><span class="tm">13:43</span></a><span class="nk"> <alicef></span> gentpatches is getting up with revision
<a href="#l-109" name="l-109"><span class="tm">13:43</span></a><span class="nk"> <alicef></span> so is not releted with the kernel version
<a href="#l-110" name="l-110"><span class="tm">13:43</span></a><span class="nk"> <mgorny></span> that's what i'd like to change
<a href="#l-111" name="l-111"><span class="tm">13:43</span></a><span class="nk"> <mgorny></span> there's rarely more than one revision for kernel version
<a href="#l-112" name="l-112"><span class="tm">13:44</span></a><span class="nk"> <mgorny></span> and it would be convenient to have both versions in sync
<a href="#l-113" name="l-113"><span class="tm">13:44</span></a><span class="nk"> <alicef></span> for example if we have gentoo-sources-5.19.10-r2 genpatches will be probably something like 12
<a href="#l-114" name="l-114"><span class="tm">13:44</span></a><span class="nk"> <alicef></span> as we have genpatches for 10 10-r1 e 10-r2
<a href="#l-115" name="l-115"><span class="tm">13:44</span></a><span class="nk"> <mgorny></span> and i'd like to see instead genpatches 10-r2 or sth like that
<a href="#l-116" name="l-116"><span class="tm">13:46</span></a><span class="nk"> <alicef></span> that's interesting
<a href="#l-117" name="l-117"><span class="tm">13:47</span></a><span class="nk"> <alicef></span> we still need the opinion of mpagano
<a href="#l-118" name="l-118"><span class="tm">13:47</span></a><span class="nk"> <mgorny></span> sure
<a href="#l-119" name="l-119"><span class="tm">13:47</span></a><span class="nk"> <alicef></span> and I think bumping genpatches with kernel version is not so straight forward as doing
<a href="#l-120" name="l-120"><span class="tm">13:47</span></a><span class="nk"> <alicef></span> n+1
<a href="#l-121" name="l-121"><span class="tm">13:48</span></a><span class="nk"> <alicef></span> we also go out of the gkernel topic :P
<a href="#l-122" name="l-122"><span class="tm">13:48</span></a><span class="nk"> <alicef></span> <span class="topic">#topic </span><span class="topicline">deblob</span>
<a href="#l-123" name="l-123"><span class="tm">13:49</span></a><span class="nk"> <mgorny></span> i think the newest deblob patch is good
<a href="#l-124" name="l-124"><span class="tm">13:49</span></a><span class="nk"> <alicef></span> about deblob I think we already had a discussion on the mailing list :)
<a href="#l-125" name="l-125"><span class="tm">13:49</span></a><span class="nk"> <alicef></span> mgorny thanks
<a href="#l-126" name="l-126"><span class="tm">13:50</span></a><span class="nk"> <alicef></span> I don't think we need furter discussion
<a href="#l-127" name="l-127"><span class="tm">13:51</span></a><span class="nk"> <alicef></span> <span class="cmd">#kernel </span><span class="cmdline">eapi mpagano whissi</span>
<a href="#l-128" name="l-128"><span class="tm">13:51</span></a><span class="nk"> <alicef></span> <span class="topic">#topic </span><span class="topicline">kernel eclass eapi mpagano whissi</span>
<a href="#l-129" name="l-129"><span class="tm">13:51</span></a><span class="nk"> <alicef></span> about kernel eapi I see mpagano that update the kernel eclass to eapi 7 and 8
<a href="#l-130" name="l-130"><span class="tm">13:51</span></a><span class="nk"> <alicef></span> still gentoo-sources are eapi 7
<a href="#l-131" name="l-131"><span class="tm">13:52</span></a><span class="nk"> <alicef></span> but I think we can alrady start to move it to eapi 8
<a href="#l-132" name="l-132"><span class="tm">13:52</span></a><span class="nk"> <alicef></span> <span class="topic">#topic </span><span class="topicline">mgorny points</span>
<a href="#l-133" name="l-133"><span class="tm">13:53</span></a><span class="nk"> <alicef></span> <span class="hi">mgorny:</span> if you have any other topic :)
<a href="#l-134" name="l-134"><span class="tm">13:53</span></a><span class="nk"> <alicef></span> but would still need mpagano opinion also
<a href="#l-135" name="l-135"><span class="tm">13:54</span></a><span class="nk"> <alicef></span> <span class="hi">montjoie:</span> anything to add about gkernelci ?
<a href="#l-136" name="l-136"><span class="tm">13:55</span></a><span class="nk"> <alicef></span> 5
<a href="#l-137" name="l-137"><span class="tm">13:56</span></a><span class="nk"> <alicef></span> 4
<a href="#l-138" name="l-138"><span class="tm">13:56</span></a><span class="nk"> <alicef></span> 3
<a href="#l-139" name="l-139"><span class="tm">13:56</span></a><span class="nk"> <alicef></span> 2
<a href="#l-140" name="l-140"><span class="tm">13:56</span></a><span class="nk"> <mgorny></span> hmm
<a href="#l-141" name="l-141"><span class="tm">13:56</span></a><span class="nk"> <mgorny></span> a rough idea that we could consider 'restarting' or merging patches in genpatches
<a href="#l-142" name="l-142"><span class="tm">13:57</span></a><span class="nk"> <mgorny></span> e.g. 4.4 series has 276 patches to bring the kernel from 4.4 to 4.4.276
<a href="#l-143" name="l-143"><span class="tm">13:57</span></a><span class="nk"> <mgorny></span> applying that many patches is slow, and they often replace one another
<a href="#l-144" name="l-144"><span class="tm">13:58</span></a><span class="nk"> <alicef></span> other distribution are usually keeping kernel sources branches and bumping that
<a href="#l-145" name="l-145"><span class="tm">13:59</span></a><span class="nk"> <mgorny></span> i think it would be more optimal to periodically merge them and make a single patch that bumps e.g. from 4.4 to 4.4.250, and then handle the rest via small patches
<a href="#l-146" name="l-146"><span class="tm">13:59</span></a><span class="nk"> <alicef></span> but is actually much more slow to keep all the kernel source code
<a href="#l-147" name="l-147"><span class="tm">13:59</span></a><span class="nk"> <alicef></span> like a script doing the periodical merge ?
<a href="#l-148" name="l-148"><span class="tm">14:00</span></a><span class="nk"> <mgorny></span> or modification to how genpatches are generated
<a href="#l-149" name="l-149"><span class="tm">14:00</span></a><span class="nk"> <alicef></span> the interesting that of keeping all patch separated is that if there is a security issue in one of kernel increment is much more fast to look for it
<a href="#l-150" name="l-150"><span class="tm">14:01</span></a><span class="nk"> <alicef></span> as we can know which patch got bumped on which kernel version
<a href="#l-151" name="l-151"><span class="tm">14:01</span></a><span class="nk"> <mgorny></span> i'm only talking of the 1* patches that backport upstream changes
<a href="#l-152" name="l-152"><span class="tm">14:01</span></a><span class="nk"> <mgorny></span> ./1000_linux-4.4.1.patch
<a href="#l-153" name="l-153"><span class="tm">14:01</span></a><span class="nk"> <mgorny></span> ./1001_linux-4.4.2.patch
<a href="#l-154" name="l-154"><span class="tm">14:01</span></a><span class="nk"> <mgorny></span> ./1002_linux-4.4.3.patch
<a href="#l-155" name="l-155"><span class="tm">14:01</span></a><span class="nk"> <mgorny></span> ./1003_linux-4.4.4.patch
<a href="#l-156" name="l-156"><span class="tm">14:01</span></a><span class="nk"> <mgorny></span> ./1004_linux-4.4.5.patch
<a href="#l-157" name="l-157"><span class="tm">14:01</span></a><span class="nk"> <mgorny></span> these ones
<a href="#l-158" name="l-158"><span class="tm">14:01</span></a><span class="nk"> <alicef></span> that's one of the pro that I can think about having it separated
<a href="#l-159" name="l-159"><span class="tm">14:01</span></a><span class="nk"> <alicef></span> yes but if for example
<a href="#l-160" name="l-160"><span class="tm">14:02</span></a><span class="nk"> <alicef></span> linux 4.4.4 have a security issue in one of the patch
<a href="#l-161" name="l-161"><span class="tm">14:03</span></a><span class="nk"> <alicef></span> sorry
<a href="#l-162" name="l-162"><span class="tm">14:04</span></a><span class="nk"> <alicef></span> I'm just think that having separate patches make more easy to search things
<a href="#l-163" name="l-163"><span class="tm">14:04</span></a><span class="nk"> <alicef></span> but is a valid point
<a href="#l-164" name="l-164"><span class="tm">14:05</span></a><span class="nk"> <alicef></span> do you think of any other benefit other than just make apply more fast ?
<a href="#l-165" name="l-165"><span class="tm">14:05</span></a><span class="nk"> <mgorny></span> smaller genpatches archive probably
<a href="#l-166" name="l-166"><span class="tm">14:05</span></a><span class="nk"> <alicef></span> I think apply last just few seconds currently
<a href="#l-167" name="l-167"><span class="tm">14:06</span></a><span class="nk"> <mgorny></span> around 10 seconds on tmpfs, i guess
<a href="#l-168" name="l-168"><span class="tm">14:06</span></a><span class="nk"> <mgorny></span> 10-15
<a href="#l-169" name="l-169"><span class="tm">14:06</span></a><span class="nk"> <mgorny></span> would be easier to measure if they were applied during src_prepare
<a href="#l-170" name="l-170"><span class="tm">14:06</span></a><span class="nk"> <alicef></span> we can add this and genpatches following genkernel version as topic for the next meeting when we have also mpagano :)
<a href="#l-171" name="l-171"><span class="tm">14:07</span></a><span class="nk"> <alicef></span> as mpagano have a much better vision on genpatches future
<a href="#l-172" name="l-172"><span class="tm">14:08</span></a><span class="nk"> <alicef></span> genpatches following kernel version and merging kernel incremental patches
<a href="#l-173" name="l-173"><span class="tm">14:09</span></a><span class="nk"> <alicef></span> I think the bot can take note
<a href="#l-174" name="l-174"><span class="tm">14:09</span></a><span class="nk"> <mgorny></span> <span class="hi">alicef:</span> if i combine the first 200 patches into one diff, genpatches goes from 4.1M to 1.2M
<a href="#l-175" name="l-175"><span class="tm">14:09</span></a><span class="nk"> <alicef></span> <span class="cmd">#note </span><span class="cmdline">genpatches following kernel version and merging kernel incremental patches</span>
<a href="#l-176" name="l-176"><span class="tm">14:10</span></a><span class="nk"> <alicef></span> <span class="cmd">#IDEA </span><span class="cmdline">genpatches following kernel version and merging kernel incremental patches</span>
<a href="#l-177" name="l-177"><span class="tm">14:10</span></a><span class="nk"> <montjoie></span> <span class="hi">alicef:</span> I have some patch to add to gkernelci, mostly adding "native build" and a way to configure workers via yaml
<a href="#l-178" name="l-178"><span class="tm">14:10</span></a><span class="nk"> <montjoie></span> they are aleady done, but I need to polish them
<a href="#l-179" name="l-179"><span class="tm">14:10</span></a><span class="nk"> <alicef></span> <span class="cmd">#idea </span><span class="cmdline">montjoie have patch to add to gkernelci, mostly adding "native build" and a way to configure workers via yaml</span>
<a href="#l-180" name="l-180"><span class="tm">14:11</span></a><span class="nk"> <alicef></span> nice !
<a href="#l-181" name="l-181"><span class="tm">14:11</span></a><span class="nk"> <alicef></span> <span class="hi">mgorny:</span> that's a nice achivement :)
<a href="#l-182" name="l-182"><span class="tm">14:12</span></a><span class="nk"> <alicef></span> let's ask mpagano opinion on the next meeting
<a href="#l-183" name="l-183"><span class="tm">14:13</span></a><span class="nk"> <alicef></span> wow configure workers via yaml would be really useful
<a href="#l-184" name="l-184"><span class="tm">14:13</span></a><span class="nk"> <alicef></span> still we need to found a way for having only one http server :/
<a href="#l-185" name="l-185"><span class="tm">14:14</span></a><span class="nk"> <montjoie></span> <span class="hi">alicef:</span> the major need is to have worker list + pass in one central file AND to permit to choose per worker which arch/toolchain to build
<a href="#l-186" name="l-186"><span class="tm">14:14</span></a><span class="nk"> <montjoie></span> for example native build for non-x86 is cpu hungry
<a href="#l-187" name="l-187"><span class="tm">14:18</span></a><span class="nk"> <alicef></span> I see
<a href="#l-188" name="l-188"><span class="tm">14:18</span></a><span class="nk"> <alicef></span> do you have any draft code ?
<a href="#l-189" name="l-189"><span class="tm">14:19</span></a><span class="nk"> <alicef></span> if you have something you should send the pull request so I can review and help out
<a href="#l-190" name="l-190"><span class="tm">14:20</span></a><span class="nk"> <alicef></span> just write draft in the title so I know is not finished yet
<a href="#l-191" name="l-191"><span class="tm">14:21</span></a><span class="nk"> <alicef></span> any other topics ?
<a href="#l-192" name="l-192"><span class="tm">14:21</span></a><span class="nk"> <alicef></span> 5
<a href="#l-193" name="l-193"><span class="tm">14:21</span></a><span class="nk"> <alicef></span> 4
<a href="#l-194" name="l-194"><span class="tm">14:21</span></a><span class="nk"> <alicef></span> 3
<a href="#l-195" name="l-195"><span class="tm">14:22</span></a><span class="nk"> <alicef></span> 2
<a href="#l-196" name="l-196"><span class="tm">14:22</span></a><span class="nk"> <alicef></span> 1
<a href="#l-197" name="l-197"><span class="tm">14:22</span></a><span class="nk"> <alicef></span> thanks mgorny
<a href="#l-198" name="l-198"><span class="tm">14:22</span></a><span class="nk"> <alicef></span> thanks montjoie
<a href="#l-199" name="l-199"><span class="tm">14:23</span></a><span class="nk"> <juippis></span> <span class="hi">alicef:</span> I have!
<a href="#l-200" name="l-200"><span class="tm">14:23</span></a><span class="nk"> <alicef></span> is already 23 minutes over let's close the meeting
<a href="#l-201" name="l-201"><span class="tm">14:23</span></a><span class="nk"> <juippis></span> it's not meeting related
<a href="#l-202" name="l-202"><span class="tm">14:23</span></a><span class="nk"> <alicef></span> juippis yes
<a href="#l-203" name="l-203"><span class="tm">14:23</span></a><span class="nk"> <alicef></span> <span class="topic">#topic </span><span class="topicline">open</span>
<a href="#l-204" name="l-204"><span class="tm">14:24</span></a><span class="nk"> <juippis></span> but since you're around, where did we left off with gkernelci github pr support?
<a href="#l-205" name="l-205"><span class="tm">14:24</span></a><span class="nk"> <juippis></span> since mgorny manages the gentoo github if there was some restriction from that side
<a href="#l-206" name="l-206"><span class="tm">14:24</span></a><span class="nk"> <alicef></span> right
<a href="#l-207" name="l-207"><span class="tm">14:24</span></a><span class="nk"> <juippis></span> now'd be a good time to talk about it
<a href="#l-208" name="l-208"><span class="tm">14:24</span></a><span class="nk"> <alicef></span> we solved that. I can now see the pr code from github
<a href="#l-209" name="l-209"><span class="tm">14:25</span></a><span class="nk"> <juippis></span> I tested it works on sys-kernel* but not on any other PRs
<a href="#l-210" name="l-210"><span class="tm">14:25</span></a><span class="nk"> <juippis></span> for regular packages
<a href="#l-211" name="l-211"><span class="tm">14:25</span></a><span class="nk"> <juippis></span> does the build server catch the request with the gkernelci label, or does it match the sys-kernel/*
<a href="#l-212" name="l-212"><span class="tm">14:25</span></a><span class="nk"> <alicef></span> yes is still not enabled for regular packages
<a href="#l-213" name="l-213"><span class="tm">14:26</span></a><span class="nk"> <alicef></span> currently the server is overloaded even for sys-kernel/*
<a href="#l-214" name="l-214"><span class="tm">14:26</span></a><span class="nk"> <juippis></span> can we enable it for regular packages with a custom label? Like gkernelci
<a href="#l-215" name="l-215"><span class="tm">14:26</span></a><span class="nk"> <juippis></span> ah
<a href="#l-216" name="l-216"><span class="tm">14:26</span></a><span class="nk"> <alicef></span> that's why I requested a dedicated machine
<a href="#l-217" name="l-217"><span class="tm">14:26</span></a><span class="nk"> <juippis></span> yes, I don't want it to test everything automatically ever. It'd be requested through a github label
<a href="#l-218" name="l-218"><span class="tm">14:26</span></a><span class="nk"> <alicef></span> sorry I have still have to work on the custom label script
<a href="#l-219" name="l-219"><span class="tm">14:26</span></a><span class="nk"> <juippis></span> that only the devs can apply
<a href="#l-220" name="l-220"><span class="tm">14:27</span></a><span class="nk"> <alicef></span> I checked it few weeks ago but I have made no progress yet
<a href="#l-221" name="l-221"><span class="tm">14:27</span></a><span class="nk"> <juippis></span> okay, it's good to know where we're at currently!
<a href="#l-222" name="l-222"><span class="tm">14:27</span></a><span class="nk"> <alicef></span> yes
<a href="#l-223" name="l-223"><span class="tm">14:27</span></a><span class="nk"> <alicef></span> gkernelci is getting each label for each pull request
<a href="#l-224" name="l-224"><span class="tm">14:28</span></a><span class="nk"> <alicef></span> currently
<a href="#l-225" name="l-225"><span class="tm">14:28</span></a><span class="nk"> <alicef></span> so gkernelci know when the gkernelci label get activated and disabled
<a href="#l-226" name="l-226"><span class="tm">14:29</span></a><span class="nk"> <alicef></span> I still need to find a way to link that to a build trigger action
<a href="#l-227" name="l-227"><span class="tm">14:29</span></a><span class="nk"> <alicef></span> I can give you the json code if you are interested
<a href="#l-228" name="l-228"><span class="tm">14:29</span></a><span class="nk"> <alicef></span> and point you to the python code related to the parsing
<a href="#l-229" name="l-229"><span class="tm">14:30</span></a><span class="nk"> <juippis></span> think I've seen them before but sure, doesn't hurt!
<a href="#l-230" name="l-230"><span class="tm">14:31</span></a><span class="nk"> <alicef></span> ok
<a href="#l-231" name="l-231"><span class="tm">14:32</span></a><span class="nk"> <alicef></span> still the current server is really slow
<a href="#l-232" name="l-232"><span class="tm">14:32</span></a><span class="nk"> <alicef></span> depend from the package it can take hour to finish :(
<a href="#l-233" name="l-233"><span class="tm">14:33</span></a><span class="nk"> <alicef></span> <span class="hi">juippis:</span> can you give me some package that are you interested to see enabled ?
<a href="#l-234" name="l-234"><span class="tm">14:33</span></a><span class="nk"> <juippis></span> yeah, that's why I don't want to enable it automatically for every PR. Someone will troll and push 100 chromium update PRs
<a href="#l-235" name="l-235"><span class="tm">14:33</span></a><span class="nk"> <alicef></span> maybe I can start from that
<a href="#l-236" name="l-236"><span class="tm">14:33</span></a><span class="nk"> <juippis></span> just pick any unmerged PR from github :P
<a href="#l-237" name="l-237"><span class="tm">14:33</span></a><span class="nk"> <alicef></span> ok :P
<a href="#l-238" name="l-238"><span class="tm">14:34</span></a><span class="nk"> <juippis></span> https://github.com/gentoo/gentoo/pull/21784 something like this
<a href="#l-239" name="l-239"><span class="tm">14:34</span></a><span class="nk"> <alicef></span> juippis thanks
<a href="#l-240" name="l-240"><span class="tm">14:34</span></a><span class="nk"> <juippis></span> but I fear someone will merge that soon. I can / you can make some dummy PR with any m-n package version bump for example
<a href="#l-241" name="l-241"><span class="tm">14:34</span></a><span class="nk"> <juippis></span> or EAPI bump
<a href="#l-242" name="l-242"><span class="tm">14:34</span></a><span class="nk"> <juippis></span> to test
<a href="#l-243" name="l-243"><span class="tm">14:35</span></a><span class="nk"> <alicef></span> yes I will do
<a href="#l-244" name="l-244"><span class="tm">14:35</span></a><span class="nk"> <juippis></span> cheers \o
<a href="#l-245" name="l-245"><span class="tm">14:35</span></a><span class="nk"> <alicef></span> thanks for the idea
<a href="#l-246" name="l-246"><span class="tm">14:36</span></a><span class="nk"> <alicef></span> I will update the issue you open in few hours with what we did
<a href="#l-247" name="l-247"><span class="tm">14:36</span></a><span class="nk"> <alicef></span> any other topic?
<a href="#l-248" name="l-248"><span class="tm">14:36</span></a><span class="nk"> <alicef></span> 5
<a href="#l-249" name="l-249"><span class="tm">14:36</span></a><span class="nk"> <alicef></span> 4
<a href="#l-250" name="l-250"><span class="tm">14:36</span></a><span class="nk"> <alicef></span> 3
<a href="#l-251" name="l-251"><span class="tm">14:36</span></a><span class="nk"> <alicef></span> 2
<a href="#l-252" name="l-252"><span class="tm">14:37</span></a><span class="nk"> <alicef></span> 1
<a href="#l-253" name="l-253"><span class="tm">14:37</span></a><span class="nk"> <alicef></span> thanks juippis mgorny montjoie
<a href="#l-254" name="l-254"><span class="tm">14:37</span></a><span class="nk"> <alicef></span> if there are no other topic I will close the meeting
<a href="#l-255" name="l-255"><span class="tm">14:38</span></a><span class="nk"> <alicef></span> <span class="cmd">#endmeeting</span><span class="cmdline"></span></pre>
</body>
</html>