Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2020-09-15

00:01:04 Quit ac_laptop (Ping timeout: 260 seconds)
00:01:09 Join t0mato [0] (~t0mato@
00:46:29***Saving seen data "./dancer.seen"
01:33:05 Join atsampson [0] (
01:41:39 Join t0mato7 [0] (~t0mato@
01:43:04 Quit t0mato (Ping timeout: 240 seconds)
01:43:05 Nick t0mato7 is now known as t0mato (~t0mato@
01:48:11 Join johnb5 [0] (
02:06:23 Quit johnb5 (Ping timeout: 258 seconds)
02:15:00 Join johnb5 [0] (
02:28:35 Quit johnb5 (Ping timeout: 240 seconds)
02:31:44 Join petur [0] (~petur@
02:31:44 Quit petur (Changing host)
02:31:44 Join petur [0] (~petur@rockbox/developer/petur)
02:46:30***Saving seen data "./dancer.seen"
03:12:50 Quit St3ak (Quit: Free ZNC ~ Powered by LunarBNC:
03:15:19 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
03:16:15 Join St3ak [0] (
03:50:04 Quit pamaury (Ping timeout: 244 seconds)
04:21:10 Join johnb5 [0] (
04:28:07 Join pamaury [0] (
04:28:07 Quit pamaury (Changing host)
04:28:07 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
04:46:33***Saving seen data "./dancer.seen"
04:57:44 Quit johnb5 (Ping timeout: 258 seconds)
05:19:03 Join johnb5 [0] (
05:26:34 Quit Acou_Bass (Ping timeout: 260 seconds)
05:31:55 Join Acou_Bass [0] (
05:32:59 Quit johnb5 (Ping timeout: 240 seconds)
06:45:04 Quit St3ak (Quit: Free ZNC ~ Powered by LunarBNC:
06:46:35***Saving seen data "./dancer.seen"
06:51:21 Join St3ak [0] (
06:53:38speachy_bilgus: Good references:, and linux/drivers/usb/musb/{jz4740.c,musbhsdma.c}
06:54:22speachyIngenic used a bone-stock MUSB HDRC implementation, along with the Inventra DMA implementation.
07:35:41 Join johnb5 [0] (
07:37:52 Quit St3ak (Quit: Free ZNC ~ Powered by LunarBNC:
07:38:19 Join St3ak [0] (
07:49:32 Quit johnb5 (Ping timeout: 256 seconds)
07:50:45 Quit St3ak (Quit: Free ZNC ~ Powered by LunarBNC:
07:51:56 Join St3ak [0] (
07:52:21 Quit St3ak (Client Quit)
07:53:44 Join St3ak [0] (
08:46:36***Saving seen data "./dancer.seen"
09:22:53speachyalmost have it. it's successfully probing the mass storage device but failing to mount
09:38:30 Quit St3ak (Quit: Free ZNC ~ Powered by LunarBNC:
09:39:31 Join St3ak [0] (
09:40:33 Quit St3ak (Client Quit)
09:41:49 Join St3ak [0] (
09:50:27 Quit St3ak (Quit: Free ZNC ~ Powered by LunarBNC:
09:52:00 Join St3ak [0] (
09:52:18 Join t0mato8 [0] (~t0mato@
09:52:38 Quit t0mato (Read error: Connection reset by peer)
09:52:38 Nick t0mato8 is now known as t0mato (~t0mato@
09:54:06 Join massiveH [0] (
09:56:32 Join amdj` [0] (~aaron@freenode/staff/atheme.amdj)
09:56:32 Quit amdj (Killed ( (Nickname regained by services)))
09:56:32 Nick amdj` is now known as amdj (~aaron@freenode/staff/atheme.amdj)
09:56:38 Quit pamaury (Quit: Konversation terminated!)
09:57:21 Join marex-cloud_ [0] (sid137234@gateway/web/
09:59:27 Join ParkerR_ [0] (
10:00:13 Quit ParkerR (Ping timeout: 272 seconds)
10:00:14 Quit marex-cloud (Ping timeout: 272 seconds)
10:00:15 Quit Rondom (Ping timeout: 272 seconds)
10:00:15 Quit WakiMiko (Ping timeout: 272 seconds)
10:00:15 Quit Tsesarevich (Ping timeout: 272 seconds)
10:01:12 Join Rondom [0] (
10:07:15 Join WakiMiko [0] (~WakiMiko@unaffiliated/wakimiko)
10:07:41 Join Tsesarevich [0] (Tsesarevic@fluxbuntu/founder/joejaxx)
10:21:09speachyok, got it! DMA is operational on the SEND side. firing up sysbench to stress it a bit..
10:30:59 Quit St3ak (Quit: Free ZNC ~ Powered by LunarBNC:
10:32:02 Join St3ak [0] (
10:34:17 Join St3ak` [0] (
10:35:17 Quit St3ak (Remote host closed the connection)
10:42:57speachy_bilgus: g#2746
10:44:31speachybenchmarks apparently unaffected. Hmm.
10:46:37***Saving seen data "./dancer.seen"
11:05:47 Quit massiveH (Quit: Leaving)
11:06:19speachythe RX side is going to be the same basic rework structure but with more complex rules.
11:11:03_bilgushaha thats funny that thing is what I was reading last night talk about serendipitous
11:18:19_bilgusspeachy very nice the removal of magic numbers helps alot
11:22:38speachyit also makes it clear what's needed to better genericize the DMA −− we can have eight different endpoints with DMA at once.
11:23:38speachyat least this driver is arguably useful beyond the jz4760, should we find another SoC that uses the same DMA engine
11:30:33 Quit dweeber (Read error: Connection reset by peer)
11:44:46_bilgusso thats odd, the reads / writes are still within that 5% range but the seq write halved in latency and Rand r/w doubled
11:45:07speachyI'd have expected no change at all in writes.
11:45:35speachysince only device->host transfers were DMA-enabled.
11:46:04speachyI do need to re-enable the logging and see what kind of transfer sizes we're actually using.
11:47:14speachy(and if mode1 DMA transfers are being used at all..)
11:48:05_bilgusI'm guessing its bypassing your code
11:52:14_bilgus((unsigned long)ptr & 4 == 0)
11:53:02speachywhoops, that should be &3
11:54:23speachyno change in sequential read.
12:09:25speachyhuh, I think it's using DMA only for the first transfer, and PIO for the remainder.
12:17:02speachyokay I must be having a major brain fart but that (ptr & 3 == 0) test is responsible for skipping dma from ever being used. but based on the log messags of the addresses used, that test is always true..
12:18:21speachy(the buffer always seems to be 16-byte aligned)
12:26:21_bilguswell removing it makes it now want to connect on usb plug
12:26:35speachyclearly badness remains.
12:34:16_bilgushmm it looks like transfer speeds drop in half after a ROLO
12:35:45_bilgusmaybe just a coincidence i'll restart and try again
12:35:51 Join ZincAlloy [0] (~Adium@2a02:8108:943f:d824:2c0a:df60:96e0:39b)
12:36:51speachyI haven't noticed that
12:44:50_bilgusnah happening after a restart too maybe somethings up with the pc
12:45:05 Quit _bilgus (Quit: Leaving)
12:46:39***Saving seen data "./dancer.seen"
12:52:51 Join _bilgus [0] (~bilgus@2605:a000:1301:89bd:846e:542c:8bed:d177)
13:08:31_bilgushmm yeah I can't get the opriginal speeds back even with the hold switch WEIRD
13:09:29 Join lebellium [0] (
13:25:20speachywith the hold switch I was seeing completely identical behavior+performance.
13:26:16_bilgusI'm going to revert to head and try again
13:27:31speachyjust pushed a new version of the patch, with some fixes but also disables DMA completely by default now.
13:31:27 Quit Huntereb (Read error: Connection reset by peer)
13:32:33_bilgusbetter than whern it started even if it does still have issues
13:33:24speachygotta put it all aside again for a while.
13:33:53_bilgusI'll play with it wed night and see what I can figure out
13:37:09_bilgusok speeds are back @ HEAD so perhaps something isn't quite right
13:37:28speachythere are zero functional changes with dma disabled
13:38:11_bilguswell you'd think the hold switch would reflect that then no?
13:38:44speachyI'd expect so, yeah. but please try with the most recent patch as the baseline (it compiles out the DMA code altogether)
13:41:01_bilgusgotta head out of town, Ill be sure to do some more testing / tweaking
13:41:32speachyI think I'll be able to poke at this again tonight. weather is crappy so it's indoor activities instead.
13:42:08speachythe linux driver is structured very differently so it's difficult to suss out some of the more subtle sequencing issues.
13:50:22 Join sakax [0] (~r0b0t@unaffiliated/r0b0t)
13:58:14 Join johnb5 [0] (
14:31:47 Join p3tur [0] (~petur@rockbox/developer/petur)
14:31:48 Quit petur (Read error: Connection reset by peer)
14:31:49 Quit p3tur (Remote host closed the connection)
14:46:41***Saving seen data "./dancer.seen"
14:51:21 Quit johnb5 (Ping timeout: 265 seconds)
15:06:44 Join johnb5 [0] (
15:20:22 Quit johnb5 (Ping timeout: 256 seconds)
15:24:40 Quit michaelni (Quit: Leaving)
15:54:20speachygot mode0 working for small transfers. big transfers (mode 1) are going into never-never land.
16:16:07 Join MrZeus_ [0] (~MrZeus@2a02:c7f:70d0:6a00:28cb:21a2:5343:b164)
16:30:40 Join michaelni [0] (
16:43:49 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
16:46:44***Saving seen data "./dancer.seen"
16:48:45speachywell. looks like the hardware doesn't support large TX transfers. Feature flag in the core says it's not implemented.
16:54:07 Join ac_laptop [0] (~ac_laptop@
16:58:45 Quit St3ak` (Quit: Free ZNC ~ Powered by LunarBNC:
16:59:41 Join St3ak [0] (
17:00:26 Quit St3ak (Client Quit)
17:03:27 Join St3ak [0] (
17:29:34 Quit michaelni (Ping timeout: 256 seconds)
17:31:01speachy_bilgus: code updated. considerably restructured.
17:35:35 Quit prg318 (Quit: ZNC 1.7.5 -
17:36:36 Quit lebellium (Quit: Leaving)
17:38:36 Join prg318 [0] (~prg@deadcodersociety/prg318)
17:41:22 Quit ZincAlloy (Quit: Leaving.)
17:42:33speachystill ahven't found why mode1 isn't working. max dma is 1MB, and the HW is suppsoed to just work. best guess is that there's some other setup bit missing.
17:42:42speachybut that's a problem for another time.
17:42:56 Quit pamaury (Ping timeout: 244 seconds)
18:02:19 Join MrZeus__ [0] (~MrZeus@2a02:c7f:70d0:6a00:d29:2940:c454:1176)
18:10:42 Quit MrZeus_ (Ping timeout: 272 seconds)
18:22:21 Quit sakax (Quit: Leaving)
18:33:15 Join Will07 [0] (
18:35:41 Part Will07
18:46:46***Saving seen data "./dancer.seen"
19:02:25 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
19:03:05 Join fs-bluebot [0] (
19:05:10 Quit fs-bluebot_ (Ping timeout: 244 seconds)
19:06:08 Quit bluebrother^ (Ping timeout: 260 seconds)
19:23:23 Quit t0mato (Ping timeout: 240 seconds)
19:23:29 Join Soap [0] (~Soap@rockbox/staff/soap)
19:49:52 Join t0mato [0] (~t0mato@
19:58:57 Quit bluebrother (Disconnected by services)
19:59:02 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
19:59:07 Join fs-bluebot_ [0] (
20:01:41 Quit fs-bluebot (Ping timeout: 272 seconds)
20:13:13speachydma transferred the first block.. and the USB core seems to be ignoring it.. kicking back an underrun error for some reason
20:36:31 Quit MrZeus__ (Ping timeout: 272 seconds)
20:46:49***Saving seen data "./dancer.seen"
21:36:22 Join Huntereb [0] (~Huntereb@
21:59:07 Join _bilgus_ [0] (~bilgus@2605:a000:1301:89bd:846e:542c:8bed:d177)
21:59:48 Quit _bilgus (Ping timeout: 244 seconds)
22:02:13speachy_bilgus_: just pushed working DMA USB TX. sequential reads jumped up about 33%, from 6.6->8.8MB/s
22:02:13 Quit _bilgus_ (Remote host closed the connection)
22:04:08 Join _bilgus_ [0] (~bilgus@2605:a000:1301:89bd:846e:542c:8bed:d177)
22:11:18speachygiven the substantial DMA overhead, it's debatable if it's worth using DMA for smaller tranfer sizes, but it's not like we're multitasking when operating as a USB device.
22:11:35 Quit ac_laptop (Ping timeout: 258 seconds)
22:46:51***Saving seen data "./dancer.seen"
23:57:31 Quit TheSeven (Ping timeout: 244 seconds)
23:57:48 Join [7] [0] (~quassel@rockbox/developer/TheSeven)

Previous day | Next day