SBo scripts not building on current (read 1st post, pls)
SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
I have problem to compile academic/ds9 on my current.
The issue could be this.
Quote:
tls.c: In function ‘get_dh2048’:
tls.c:121:7: error: dereferencing pointer to incomplete type ‘DH’ {aka ‘struct dh_st’}
dh->p=BN_bin2bn(dh2048_p,sizeof(dh2048_p),NULL);
^~
tls.c: At top level:
tls.c:171:24: error: ‘CRYPTO_NUM_LOCKS’ undeclared here (not in a function); did you mean ‘CRYPTO_UNLOCK’?
static Tcl_Mutex locks[CRYPTO_NUM_LOCKS];
I'm on a current distro, with blacklisted kernel modules at 4.14.67 .
Additionally, the maintainer cannot be reached, because his email is registered with the same server, which should hold the archive, but the server is down.
The original domain name was `slaphappygeeks.com', which can be resolved, but the server doesn't work.
Additionally, the maintainer cannot be reached, because his email is registered with the same server, which should hold the archive, but the server is down.
My bad - I am (was) the maintainer and I have long been intending to get that moved or back online. Among my other todo's is to catch up with my few SBo scripts or pass them to someone currently more able to maintain them.
However, as I was self-hosting the snapshot at one time, I realised that it might disappear at some poiint, so I added a backup method to the Slackbuild script - look in the comments near the top.
There are instructions there that will allow you to grab and build a current snapshot from the apache project servers.
Code:
mkdir svn
cd svn
svn co http://svn.apache.org/repos/asf/xalan/c/trunk
You can then tar that up using the svn version number (or other name) and change the version in the Slackbuild script to use that copy instead.
If that doesn't work for you please send me a PM and I'll get you a copy of the archive mentioned in the script.
My bad - I am (was) the maintainer and I have long been intending to get that moved or back online. Among my other todo's is to catch up with my few SBo scripts or pass them to someone currently more able to maintain them.
However, as I was self-hosting the snapshot at one time, I realised that it might disappear at some poiint, so I added a backup method to the Slackbuild script - look in the comments near the top.
There are instructions there that will allow you to grab and build a current snapshot from the apache project servers.
Code:
mkdir svn
cd svn
svn co http://svn.apache.org/repos/asf/xalan/c/trunk
You can then tar that up using the svn version number (or other name) and change the version in the Slackbuild script to use that copy instead.
If that doesn't work for you please send me a PM and I'll get you a copy of the archive mentioned in the script.
Thanks! Maybe I can try to write a script to automagically extract the commit and version number from the git repo .
OpenCV automatic dependency finder cannot build opencv with ATLAS.
The reason, perhaps, being the lapacke/clapack.
Indeed, I did manage to compile lapacke from academic/lapacke, and it indeed can pick up ATLAS as a systemwide LAPACK, but the resulting lapacke is not satisfactory for opencv to accept.
I suspect that lapacke should be build by the atlas builder itself, while building ATLAS, but I didn't find where to ask it politely enough.
(i.e. I haven't found a place where to give parameters with which ATLAS would build lapack)
OpenCV automatic dependency finder cannot build opencv with ATLAS.
The reason, perhaps, being the lapacke/clapack.
Indeed, I did manage to compile lapacke from academic/lapacke, and it indeed can pick up ATLAS as a systemwide LAPACK, but the resulting lapacke is not satisfactory for opencv to accept.
I suspect that lapacke should be build by the atlas builder itself, while building ATLAS, but I didn't find where to ask it politely enough.
(i.e. I haven't found a place where to give parameters with which ATLAS would build lapack)
Any suggestions welcome...
Hi Lockywolf,
is this behaviour specific to current? if you can confirm it happens also on stable or if you are just not sure that it happens only on current (like explained in the first post of this topic), you might have a better help posting it to the slackbuilds-users mailing list
I have problem to compile academic/ds9 on my current.
The issue could be this.
I'm on a current distro, with blacklisted kernel modules at 4.14.67 .
hi simotrone,
I tested this and building against an older version of openssl looks mandatory: I fiddled with it a while but unfortunately I hadn't been able to find a way to do it correctly
try to get in touch with the maintainer, maybe he could help!
hy qunying, I just tested and inkscape from the repository seems to build fine here on slackware64-current...
note that I already apply two patches for the newer poppler
I don't seem to be able to compile opencolorio and openimageio. (My final aim being Blender.)
I tried checking out the latest versions from git, but those also don't seem to make it, even though the errors are different.
Any suggestions would be greatly welcome...
I'm not sure what's happening there but I just tried to build them here (opencolorio and openimageio) on slackware64-current and I can confirm you that they both build fine.
sorry if I ask, have you read the first post of this topic?
Hmm, not sure if I should email maintainers, or just post it here...
After the recent kernel 4.19.7, nvidia-driver we have in the repo broke.
I thought that updating the driver will be more of a hassle, since it consists of several packages, and moreover, before updating nvidia-driver, one must not forget to run nvidia-switch (which I do all the time).
So I patched the old nvidia-kernel SlackBuild 390.59 myself and relaxed.
If someone needs the patch, I just added the following lines to the SlackBuild, after the line 74:
Code:
sed -i s/drm_mode_connector_update_edid_property/drm_connector_update_edid_property/g kernel/nvidia-drm/nvidia-drm-connector.c
sed -i s/DRM_CONTROL_ALLOW/00000/g kernel/nvidia-drm/nvidia-drm-drv.c
sed -i s/drm_mode_connector_attach_encoder/drm_connector_attach_encoder/g kernel/nvidia-drm/nvidia-drm-encoder.c
These functions were renamed in the recent kernel.
I also managed to fight the opencv dependency resolution failure.
Again, the patch by no means guarantees that opencv built like this will work correctly, but on my machine it managed to get compiled.
So:
1)In the lapack-atlas, I replaced the address of lapack-3.6.1 with the one of the latest lapack 3.8.0, and reinstalled the package.
3)I reinstalled ATLAS, which now had the latest lapack 3.8.0.
4)I reinstalled lapacke, which also is 3.8.0
5)I patched the opencv dependency resolution system.
That is, I added the following lines to the SlackBuild after the line 94:
Code:
sed -i '54 a set(LAPACK_LIBRARIES "${LAPACK_LIBRRARIES};/usr/lib64/liblapack.a;/usr/lib64/libptcblas.a;/usr/lib64/libf77blas.a;/usr/lib64/libatlas.a;/usr/lib64/libgfortran.a;/usr/lib64/libquadmath.a")' ../cmake/OpenCVFindLAPACK.cmake
And reinstalled opencv. (OpenCV's resolution system is poor, also, it downloads caffe models even if no caffe is installed, but I have cheap traffic, so I didn't care.)
Lockywolf if this about Blender, the binary version they provide works great if you plan to use CUDA. NVIDIA's CUDA developers only support Red Hat, SUSE, and Ubuntu so you are stuck with whatever gcc they use. The binary provides working CUDA.
If I remember correctly opencolorio won't build if openimageio is installed.
Opencolorio and openimageio have a mutual dependence, which is a bit of a pain, but luckily, the depth of the recursion it's only about 2, so it's possible to build one of them without the other, then the second, then the first again, thus enabling the recursion, and then the second again.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.