Testing some - perhaps many - cams with CHDKPTP - page 8 - General Discussion and Assistance

reyalp

14080

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #70 on: 03 / November / 2019, 21:52:08 »

Advertisements

@koshy Here's another test build for ixus30

ixus30_sd200-100k-1.5.0-5289-reyalp-ptp-test-2.zip (855.22 kB - downloaded 8 times.)

Logged

Don't forget what the H stands for.

koshy

1096

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #71 on: 05 / November / 2019, 09:58:39 »

Quote from: reyalp on 03 / November / 2019, 21:52:08

@koshy Here's another test build for ixus30

Cam crashes but "Romlog could not be saved". Tried both on the NC10 and due to the slight XP-phobia on the main workstation with the extenders as well, same crash. Just didn't want to start off with an OS specific issue, so...

i30_II.txt (1.13 kB - downloaded 8 times.)

Logged

Koshy had a little ELPH which wasn't white as snow but everywhere that Koshy went the ELPH was sure to go. (actually an SD, but that detail ruins the rhyme...)

reyalp

14080

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #72 on: 05 / November / 2019, 12:34:02 »

Quote from: koshy on 05 / November / 2019, 09:58:39

Quote from: reyalp on 03 / November / 2019, 21:52:08
@koshy Here's another test build for ixus30
Cam crashes but "Romlog could not be saved". Tried both on the NC10 and due to the slight XP-phobia on the main workstation with the extenders as well, same crash. Just didn't want to start off with an OS specific issue, so...

Looks like it crashes as soon as it tries to do anything over PTP? Probably a mistake in my code in that case.

Background:
This is one of the cameras that fails on 52 byte transfers.
My previous test broke the 52 byte transfers into a 48 and 4 byte recv calls, which didn't help
The last test was supposed to try the same thing using Canon native PTP buffers (=uncached memory) but the usual functions aren't identified by the sig finder (in fact, the size function doesn't exist, the size is hard coded in the Canon code)

Logged

Don't forget what the H stands for.

koshy

1096

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #73 on: 05 / November / 2019, 13:40:19 »

Quote from: reyalp on 05 / November / 2019, 12:34:02

Looks like it crashes as soon as it tries to do anything over PTP?

Yes, with the CHDKPTP GUI I can connect, can even get the Liveview to transfer the current image being displayed ("Viewfinder") but tick UI Overlay and boom, or try to switch to Rec mode, boom, etc. So, as soon as we "do" pretty much anything the cam crashes, no ROMLOGs.

Logged

Koshy had a little ELPH which wasn't white as snow but everywhere that Koshy went the ELPH was sure to go. (actually an SD, but that detail ruins the rhyme...)

srsa_4c

4451

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #74 on: 05 / November / 2019, 13:50:40 »

Quote from: koshy on 05 / November / 2019, 13:40:19

try to switch to Rec mode, boom

FWIW, that is highly unlikely to work on this camera, regardless of the current PTP experiment.

Logged

koshy

1096

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #75 on: 05 / November / 2019, 14:12:41 »

Quote from: srsa_4c on 05 / November / 2019, 13:50:40

Quote from: koshy on 05 / November / 2019, 13:40:19
try to switch to Rec mode, boom
FWIW, that is highly unlikely to work on this camera, regardless of the current PTP experiment.

We'll take the tests however far reyalp wants to, it sure is an antique as far as CHDK-Cams go...

Logged

Koshy had a little ELPH which wasn't white as snow but everywhere that Koshy went the ELPH was sure to go. (actually an SD, but that detail ruins the rhyme...)

koshy

1096

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #76 on: 06 / November / 2019, 19:54:27 »

Quote from: koshy on 30 / October / 2019, 11:56:41

Quote from: reyalp on 29 / September / 2019, 21:22:44
Quote
What cams shall be re-tested from the ones we already looked at?
The only one that comes to mind is ixus200_sd980, using a trunk build (not the test build I posted earlier)
This camera failed the bug 0x1f5 test with the trunk build, which was apparently fixed by enabling the bug 0x23f4 fix.
This turned out to be interesting. This is the first test done on the Samsung NC10 Win XP with plain vanilla USB 2.0 ports test system I now have on my desk. All subsequent posts will be from that one unless specifically marked otherwise. What happened was there was no problem. Just forcing the issue by looping the test did lead to a crash eventually - on the 7th iteration. So, that it passes once does not seem to be generalizable...

Quote from: reyalp on 02 / November / 2019, 23:07:39

The crash was in shooting related code, so it's not the 0x1f5 size bug we saw on the previous USB setup. The MotionVector.c assert may indicate an issue with hook placement we could fix in this port. A few ports have this. It's good to note, but not directly related to the USB issues.

So, I tried a trunk build of i200 from a time when the trunk should have suffered from the 0x1f5 size bug and while the cam crashed on something urelated 7 iterations were unaffected by the bug being sought on the XP

(from the expression on that face one would think I had said CP/M) test system. So, what does that mean? That I should re-test the i200 on various machines (excluding the CP/M ones

) so that we may puzzle out where the 0x1f5 size bug does kick in?

Logged

Koshy had a little ELPH which wasn't white as snow but everywhere that Koshy went the ELPH was sure to go. (actually an SD, but that detail ruins the rhyme...)

reyalp

14080

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #77 on: 06 / November / 2019, 20:12:23 »

Quote from: koshy on 06 / November / 2019, 19:54:27

So, I tried a trunk build of i200 from a time when the trunk should have suffered from the 0x1f5 size bug and while the cam crashed on something urelated 7 iterations were unaffected by the bug being sought on the XP (from the expression on that face one would think I had said CP/M) test system. So, what does that mean? That I should re-test the i200 on various machines (excluding the CP/M ones ) so that we may puzzle out where the 0x1f5 size bug does kick in?

The original 0x1f5 bug should be fixed on all cameras in the trunk, since r5235 on August 4.

If I've kept my notes correctly, ixus200 build with this fix failed the 0x1f5 test when you were using the extender, and passed without.

If this is correct, no further testing of ixus200 is needed. I might look at the MotionVector thing at some point, but there's nothing to test related to it now.

Logged

Don't forget what the H stands for.

koshy

1096

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #78 on: 07 / November / 2019, 06:44:51 »

Quote from: reyalp on 06 / November / 2019, 20:12:23

The original 0x1f5 bug should be fixed on all cameras in the trunk, since r5235 on August 4.

By chance exactly what I used, r5235. Thanks, that makes sense of it.

Quote from: reyalp on 06 / November / 2019, 20:12:23

If I've kept my notes correctly, ixus200 build with this fix failed the 0x1f5 test when you were using the extender, and passed without.

Yes with the USB 2 gadget things turned out to be flaky in that regard. I still haven't figured out make/model of the thing...

Quote from: reyalp on 06 / November / 2019, 20:12:23

If this is correct, no further testing of ixus200 is needed.

That's good, so we'll just go on with untested cameras.

Logged

Koshy had a little ELPH which wasn't white as snow but everywhere that Koshy went the ELPH was sure to go. (actually an SD, but that detail ruins the rhyme...)

koshy

1096

Re: Testing some - perhaps many - cams with CHDKPTP

« Reply #79 on: 08 / November / 2019, 19:19:17 »

Tested my remaining 7 SX series cams tonight.
All passed 25/25 tests except the SX200.

sx.zip (9.62 kB - downloaded 8 times.)

Logged

Koshy had a little ELPH which wasn't white as snow but everywhere that Koshy went the ELPH was sure to go. (actually an SD, but that detail ruins the rhyme...)