-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
sci-libs/andor-camera-driver-3.7.30008.0 fails to compile with sys-kernel/linux-headers-3.4 due to deprecated use of create_proc_read_entry and VM_RESERVED. #23
Comments
FYI, I replaced vm_reserved with vm_dontexpand | vm_dontdump in order to build on 3.9 here, without side effects. One thing to note: according to Norman with Andor 3rd party support, sdk3 does not support events, for example buffer overflow and exposure. I think he's referring to feature events, and the lack of support is observable in that, at least in my testing, any feature event callback is called immediately upon registration. On Feb 23, 2014, at 10:42 AM, Pariksheet Nanda notifications@github.com wrote:
|
Thanks! I've applied your vm_reserved fix and eselect'ed my previous 3.8 kernel On 02/23/2014 10:28 PM, erikhvatum wrote:
|
I did some experimenting with feature events and looked carefully at Regarding feature events, the manual states, "As soon as this callback So, AndorSDK3 will notify you when a change to a feature that you I am doing further experimentation to determine what impact this has On Mon, Feb 24, 2014 at 5:29 AM, Pariksheet Nanda
|
Did you finally manage to compile the Bitflow driver on 3.1x kernels? I have a Neo which is basically unusable on any recent distributions. |
@matze looks like we are in luck: Andor posted an updated Linux SDK to MyAndor this morning. I downloaded it, and it builds against kernel 3.9.11 without modification, which is already an improvement over the old SDK. I'm about to head home for the weekend, so I'm reluctant to upgrade the kernel on our camera system at the moment :) I'll give it a try on Monday, however. |
Great news! I'll also try the new SDK on Monday. |
Crap... the new SDK still has bitflow code that is trying to use create_proc_read_entry, which doesn't exist is 3.10+. I'm reading through the updated procfs docs and attempting to make the necessary changes to blkfi.c for 3.10+... Update: the changes needed require more thought than I have time for at the moment. It's not just a matter of replacing the create_proc_read_entry call with create_proc_data, which allows associating a single void* user data blob. There is also an array of character arrays that comes from somewhere. Putting a lot of printks in there will help figure this out, but in any case requires running the driver where it works unmodified. So, I'm back on 3.9.11 with the latest SDK installed. |
I don't have Andor's approval yet but that's a pity :-/ Looking forward to your attempt. |
proc_fs
removedcreate_proc_read_entry
and instead suggest usingproc_create[_data]
.VM_IO
,VM_DONTEXPAND
,VM_HUGETLB
, andVM_PFNMAP
to replaceVM_RESERVED
http://article.gmane.org/gmane.linux.kernel.mm/83212/Andor's bundled version of the BitFlow kernel driver often lags behind upstream, but BitFlow does not publically list their Linux SDK. Therefore BitFlow has been contracted to see if patches are available.
The text was updated successfully, but these errors were encountered: