Is it possible to compile NVIDIA for multiple kernels simultaneously?

Hi all,
First, the reference is http://wiki.archlinux.org/index.php/NVI … tom_kernel, as well as a small discussion I started at http://aur.archlinux.org/packages.php?ID=20410.
Basically, currently the nvidia-beta (and I'm sure the rest of the nvidia scripts) use `uname -r` to obtain the current kernel name, which is then used for the rest of the PKGBUILD. Its relatively trivial to alter that to the name of whichever kernel, meaning there's no problem compiling nvidia.ko for kernel26-ice from within kernel26 itself, and to alter the name of the package so that your nvidia-beta package does not need to be over-written (you then have two nvidia.ko, each for their own kernel.
My question is, is it possible for a single PKGBUILD to compile nvidia.ko for all kernels currently installed (probably through looking at either /lib/modules or /usr/src)? This would mean only one maintainer is needed for nvidia-beta or nvidia-17xx or whichever series, and the same PKGBUILD could be used on any system, even those with multiple kernels.
I'm not very familiar with the limitations of PKGBUILDs, or the rules governing what they may or may not do (for example, accessing and checking the machine itself outside the fakeroot?), so this may be nonsense, if so feel free to point it out. Another idea would be for a simple bash script to create as many PKGBUILDs as required, based on the most recent nvidia-beta PKGBUILD, simply sed-replacing uname -r with the appropriate kernel names and modifying the PKG name.

Hi,
it's not difficult to do but I DO NOT recommend to do it (see my notes below the code).
PKGBUILD
pkgname=nvidia-beta
pkgver=185.18.04
pkgrel=1
pkgdesc="NVIDIA beta drivers for kernel26."
arch=('i686' 'x86_64')
[ "$CARCH" = "i686" ] && ARCH=x86 && NV=0
[ "$CARCH" = "x86_64" ] && ARCH=x86_64 && NV=0
provides=(nvidia=${pkgver})
url="http://www.nvidia.com/"
depends=(kernel26 nvidia-utils-beta)
conflicts=('nvidia-96xx' 'nvidia-71xx' 'nvidia-legacy')
license=('custom')
install=('nvidia.install')
source=(ftp://download.nvidia.com/XFree86/Linux-$ARCH/${pkgver}/NVIDIA-Linux-$ARCH-${pkgver}-pkg${NV}.run)
md5sums=('95b67e6f17cc54d36ae2743df76c9f6d')
[ "$CARCH" = "x86_64" ] && md5sums=('33967a6d452c562bb8af02b0734ebf22')
build()
# Read list of installed kernels
KERNELS=`ls -1 /lib/modules`
# Extract the nvidia drivers
cd $startdir/src/
sh NVIDIA-Linux-${ARCH}-${pkgver}-pkg${NV}.run --extract-only
cd NVIDIA-Linux-${ARCH}-${pkgver}-pkg${NV}
cd usr/src/nv
ln -s Makefile.kbuild Makefile
# Now build modules for each kernel
for KERNEL in $KERNELS; do
echo "Building for ${KERNEL}"
# Compile the module
make SYSSRC=/lib/modules/${KERNEL}/build module || return 1
# Install kernel module
mkdir -p $startdir/pkg/lib/modules/${KERNEL}/kernel/drivers/video/
install -m644 nvidia.ko $startdir/pkg/lib/modules/${KERNEL}/kernel/drivers/video/ || return 1
# Clean
make clean || return 1
done;
nvidia.install
# arg 1: the new package version
post_install() {
KERNELS=`ls -1 /lib/modules`
for KERNEL in $KERNELS; do
depmod -v $KERNEL > /dev/null 2>&1
done;
# arg 1: the new package version
# arg 2: the old package version
post_upgrade() {
post_install $1
rmmod nvidia || echo 'In order to use the new nvidia module, exit Xserver and unload it manually.'
# arg 1: the old package version
post_remove() {
KERNELS=`ls -1 /lib/modules`
for KERNEL in $KERNELS; do
depmod -v $KERNEL > /dev/null 2>&1
done;
op=$1
shift
$op $*
PS: I haven't test it because I don't use multiple kernels
But be aware of possible problems. Some modifications may influence kernel functionalities, so once you compile a driver for kernel26-xyz on kernel26-zyx the module may not work on kernel26-xyz or may cause unexpected issues.
You also should not (you possibly even won't be able to) build module for different release version (eg. build for 2.6.28 on 2.6.29) due changes in kernel headers.
I'm also not sure if it is safe to generate module dependencies for another kernel then you are currently running.
As you can see above, it is not problem to make such a PKGBUILD, but it also brings too many problems and it's too dangerous that it is better to stay at current model "module package for each kernel". So if you want to you may try to use this PKGBUILD, but I strictly warn everyone else that it may cause more harm then use so don't blame me if something goes wrong. You were warned .
Last edited by ProgDan (2009-04-30 13:48:06)

Similar Messages

  • Can't change the 'compilation status' for multiple items in iTunes 8.

    Hi there,
    I've got an issue regarding the disappearance of the compilation 'tick box' when changing multiple items. Its there when you click on 'get info' for the individual track but its no longer there (in iTunes 8) when I want to change the compilation 'status' for multiple tracks at a time.
    In iTunes 7 I selected all the tracks and changed their compilation status by accident, but now, in iTunes 8 I can't change this without going through my library, track by track!.
    Please help.
    Thanks,
    Fred

    Multiple Get info now has tabs - look on the options tab.

  • Can I search for multiple images simultaneously in Aperture?

    Can I search for multiple images simultaneously in Aperture?  For example, search for 'Image2.jpg, Image3.jpg, Image5.jpg' by image name, out of different folders? 

    dusan -- your best procedure is likely to be to type the first file name (without extension) in the search field of the Photos View, and then mark that file either with a Flag or a color label.  (You might clear all Flags prior.)  If you have a lot of images or a slow computer (or both ) you may have to wait a few seconds.  Aperture will show "Loading ... " on the left side of the tool strip while it churns through your images.
    Then hit backspace twice, and type "60".  Mark the file found.
    Repeat with variations.
    Then go to Flagged view, or filter the whole Photos view for the color-label you used.

  • Is it possible to amend metadata for multiple selection

    i recently "updated' to Photos, the jury is still out as to whether it is an improvement or not.  I now have huge blocks of photos in the wrong timeline and need to adjust the metadata date.  I know it's possible to amend a single photo but cannot do this for multiples and yet when I click the date on a single file the pop up window suggests that it impossible as it states "adjust time for selected photos" yet i am unable to get this option.

    You can adjust the date and time of several photos from the menu "Image > Adjust Date &Time".
    That will shift the time on all selected photos by the same amount of time:
    What Photos cannot do, is to set the date to the same time for all photos at once, like iPhotos Batch Change command could do.

  • Nouveau for multiple kernels?

    Hi all,
    Am thinking of trying out the nouveau driver, since I've been gaming a LOT less lately. I understand I need xf86-video-nouveau (roughly analogous to the nvidia-utils package), nouveau-firmware, and nouveau-drm (roughly analogous to the nvidia package).
    Thing is, I use 3 kernels and switch between them fairly regularly. Because of this I use (and maintain) the nvidia-beta-all package on the AUR. In order to get this same level of functionality with nouveau I'd need something like nouveau-drm-all.
    There is an nouveau-drm-bfs PKGBUILD in the AUR, but all that does is copy the .ko files built for 2.6.33-ARCH to a new /lib/modules/* location. I'm pretty sure that wouldn't work for all kernels (especially the -rt kernel I'm fond of using), so I'd like to ask those who are already using nouveau, do you use custom kernels? Does re-running the nouveau-drm PKGBUILD on your custom kernel work?

    Actually, more needs to be done, since the PKGBUILD I have in mind will build N copies of nouveau-drm .ko files if you have N kernels installed. I've never understood, for example, why nvidia-beta is written in such a way that if I use it in my custom kernel and for some reason can't boot the custom kernel, I don't have nvidia.ko in my stock -ARCH kernel so I can't even get GUI on that.
    In any case, I believe I would also have to modify the Makefile, which uses `uname -r` in several places. nouveao-drm-bfs doesn't, because that PKGBUILD writer assumed that you would run it while kernel26-bfs is already installed and running. My nvidia-beta-all PKGBUILD can be used whatever kernel happens to be booted at the time.

  • Show file size in Cell Extras? (Either Compact or Expanded, or anywhere in Grid for multiple images simultaneously)

    I've searched and seen several suggestions saying to change Metadata to Exif & IPTC to see the file size, but that's for only one image at a time. Under Cell Extras (both Compact and Expanded) I see options for Megapixeps, Cropped Dimensions, File Name, Extention, and a bunch of other options, but no file size. I suspect this is because file size isn't a metadata field and simply a file attribute (that may be incorrect, but regardless...) is there really no way to have file size visible in the thumbnail cells in Grid view in the Library? I'm trying to reduce redundancy and viewing file size of multiple images at a glance based on my Library Filter criteria would be a huge help. Thanks to anyone who can help sort me out!

    Sure. I have a catalog with 22,000 images indiscriminately imported from hundreds of folders and multiple drives from this year alone (partially lost a RAID array to a corrupt index and recovered a dump of the files so I'm terribly disorganized juggling emergency redundancy backups wherever I had 50 gigs here, 100 gigs there, and working to clean up and organize my photos one year at a time ), including raw, jpg, websized jpg, png, tiff and psd (flattened and with layers/adjustments) and am now trying to identify my keepers and delete unnecessary files. I have 22 terabytes of personal files and photography spread chaotically and while I now have proper discipline when saving variants, I didn't always and want to remove unneeded, rejected, and duplicate images. I don't expect to be caught up for months or longer depending on my free time and sanity. Unfortunately, it's not as simple as keep the raw delete the rest. Sometimes I have multiple PSD documents with different version names and edits, sometimes with multiple layers, and I have to use a variety of information to identify which one is indeed the correct one to keep...or open them all/both.
    Often, file size can be an indicator for me and aids in making the decision more quickly. For instance between 2 psd files with the same Megapixles I'm often able to identify the flattened version vs the layers intact version by file size if it wasn't properly noted in the filename, which unfortunately due to a combination of bad or evolving practices and being tired or rushed happens more often than I'd like to admit. One I just came across that brought me here searching was 3 similarly named and appearing PSD files of a panorama, all with the same MP/dimensions. One was flattened, one was the original stitch, and one was flattened with a few masks, adjustments, and an extra layer, and being the one I'd want to keep. Filesize helped me quickly identify which was which. Also sometimes dimensions are misleading, like when a panorama is first created and uncropped, its dimensions are larger, but a tighter cropped image with multiple layers and a larger file size indicates to me that it is newer and more finished, even though smaller dimensions typically indicates a crop and may not be a 'master' copy. Additionally, 2 images with different names but the exact same file size are very likely duplicates (at least in my workflow) but 2 similarly looking images with the same dimensions could be retouched/not, sharpened/not, have noise reduction/not, and I'd have to view them in detail to determine if they were in fact identical. For my purposes (since these are my personal fine art images & I have the raw files incase I occasionally mess up) I'm often willing to trade an educated guess for my time. Seeing identical file sizes on 2 apparently similar images is good enough for me to delete one.
    I can select them individually and check...it's not a horrific problem. But it would save me time so I was hoping I had simply missed how to view that. My reasoning is 'why not' have the option available with all the others? Me personally, I don't care about megapixels when I already have the dimensions, that's just redundant and honestly much less helpful than LxW. File size would be an additional tool I would use, though. There are other examples but the main thing is that I do spend time checking filesize, and specifically when comparing multiple images, so having it displayed in the cell would be a time saver for me.

  • Listening for multiple keypresses simultaneously

    I'm trying to move a rectangle around with arrow keys.
    The current code of keyPressed method is
         public void keyPressed(KeyEvent e){
    /*           switch(e.getKeyCode()){
                   case 37: x=(Math.max(0,x-3));break;
                   case 38: y=Math.max(0,y-3); break;
                   case 39: x=Math.min(200,x+3);break;
                   case 40: y=Math.min(200,y+3);break;
              int a = e.getKeyCode();
              if(a == 37){x=Math.max(0,x-3);}
              if(a == 38){y=Math.max(0,y-3);}
              if(a == 39){x=Math.min(200,x+3);}
              if(a == 40){y=Math.min(200,y+3);}
              repaint();
         }The problem is that when pressing multiple keys at the same time only one registers.
    So how do I register multiple key presses ad change x and y accordingly?
    thank you

    It seems to me that the problems mentioned in that link with the "hold" could be solved with using a state flag set on keyPressed and then reset on keyReleased. When you release even one you should get the keyReleased event and until then, you should be in a hold state.

  • How to make Adobe acrobat feature to convert SAP  Pages to PDF available for multiple users connected to the same Citrix server

    Hi,
    In my previous endeavours to solve this business requirement where multiple users will be able to use the Adobe acrobat feature to convert SAP pages inside SAP to PDF, I was told that it is not possible to do this for multiple users at the same time. However I have found an article according to which it says it is possible. Could you check it once and let me know if this article can be used for implementing the requirement stated above as this link clearly indicates that Adobe Acrobat is supported on Citrix.
    Please find the link below where it states it is adobe acrobat is supported in Citrix for multiple users.
    http://www.adobe.com/devnet-docs/acrobatetk/tools/AdminGuide/citrix.html.

    1. What is the Acrobat feature that you mean? There's nothing specific to SAP included with Acrobat.
    2. Yes, some Terminal Server configurations appear supported (check carefully). All users of the terminal server will need a license, so far as I know - total licenses = total individual people using.

  • Create confirmation for multiple cost objects PO in SUS

    Hi,
    Is it possible to create confirmation for multiple cost objects PO ( order type 'MA') in SUS ?
    We are going to implement SRM 7.
    Thanks & regards,
    Afandi-

    Hi All,
    First of all, thanks for your responses..
    Right, about the issue, what I explained here was that I am indeed assigning only one cost object: the WBS element. The issue was that even though I am assigning only the WBS element, it was also assigning the cost to the cost center by default. I did some R&D and found the solution to the issue (I was also asked to look for OSS notes but was not satisfied that this issue needs an OSS note to be applied so tried my solution). The issue was in table: T788M (allocate costs to variable account assignment). Here, I created an entry and called it USERGROUP_2 (just a random name) and assigned the variable cost objects (only the WBS and the Cost center) to be displayed. In the next step, I assigned this usergroup to the country in quesion feature (TRVCO). By doing this, I tell the system that only these cost objects are to be considered when an employee wants to assign the cost object. If the system sees that there is no value from the drop down to choose from, it picks up the default cost object (cost center). This was a simple issue that I had to rattle my brains on... but the solution I mention above worked like a hot knife going through butter...
    If you guys face this issue, please try this else feel free to get in touch with me on my number below.
    Once again, thanks for your responses.
    Best regards,
    Tanmay Dhingra
    +91 880 6666 606

  • Execution of Task for multiple posting period

    Dear Experts,
    As i have seen it is compulsory to punch posting Period value in Permanent Parametere, Is there any possibility to exeucte Task for multiple period at a time.
    Thanks
    Ritesh M

    Yes, there is such a possibility. But it requires EhP2:
    http://help.sap.com/erp2005_ehp_02/helpdata/en/45/2b650d4ad46d95e10000000a114a6b/frameset.htm

  • Variant for multiple reports

    Hi Experts,
    Does anyone know if it's possible to create variants that serve for multiple reports?  If so, how?
    I have 10 very similar reports which use the same selection screen defined in an include for all of them.
    Regards,
    Simon

    Hi,
    It is not possible to create variant for multiple reports as it is said as
    you must create a separate variant for each report and then save it with the report where the values will be used.
    refer this line in the link
    http://web.mit.edu/SAPR3/docs/webdocs/reports/rpRFvariant.html

  • Compiled Driver for Nvidia Nforce

    Hello,
    Does anyone know where we can find a compiled Linux driver for the Nvidia
    Nforce ethernet controller. We are trying to image new computers. We
    have downloaded the drivers from the Nvidia website, but are unable to
    extract/install the files from the file. Is it possible to compile the
    driver/file from a computer running Windows?
    Any help would be appreciated
    Thanks
    Pat

    Pjsuko,
    It appears that in the past few days you have not received a response to your
    posting. That concerns us, and has triggered this automated reply.
    Has your problem been resolved? If not, you might try one of the following options:
    - Do a search of our knowledgebase at http://support.novell.com/search/kb_index.jsp
    - Check all of the other support tools and options available at
    http://support.novell.com.
    - You could also try posting your message again. Make sure it is posted in the
    correct newsgroup. (http://support.novell.com/forums)
    Be sure to read the forum FAQ about what to expect in the way of responses:
    http://support.novell.com/forums/faq_general.html
    If this is a reply to a duplicate posting, please ignore and accept our apologies
    and rest assured we will issue a stern reprimand to our posting bot.
    Good luck!
    Your Novell Product Support Forums Team
    http://support.novell.com/forums/

  • Is it possible to use single ssl certificate for multiple server farm with different FQDN?

    Hi
    We generated the CSR request for versign secure site pro certificate
    /* Style Definitions */
    table.MsoNormalTable
    {mso-style-name:"Table Normal";
    mso-tstyle-rowband-size:0;
    mso-tstyle-colband-size:0;
    mso-style-noshow:yes;
    mso-style-priority:99;
    mso-style-qformat:yes;
    mso-style-parent:"";
    mso-padding-alt:0in 5.4pt 0in 5.4pt;
    mso-para-margin:0in;
    mso-para-margin-bottom:.0001pt;
    mso-pagination:widow-orphan;
    font-size:11.0pt;
    font-family:"Calibri","sans-serif";
    mso-ascii-font-family:Calibri;
    mso-ascii-theme-font:minor-latin;
    mso-fareast-font-family:"Times New Roman";
    mso-fareast-theme-font:minor-fareast;
    mso-hansi-font-family:Calibri;
    mso-hansi-theme-font:minor-latin;}
    SSL Certificate for cn=abc.com   considering abc.com as our major domain. now we have servers in this domain like    www.abc.com,   a.abc.com , b.abc.com etc. we installed the verisign certificate and configured ACE-20 accordingly for ssl-proxy and we will use same certificate gerated for abc.com for all servers like www.abc.com , a.abc.com , b.abc.com etc. Now when we are trying to access https//www..abc.com or https://a.abc.com through mozilla , we are able to access the service but we are getting this message in certfucate status " you are connected to abc.com which is run by unknown "
    And the same message when trying to access https://www.abc.com from Google Chrome.
    "This is probably not the site you are looking for! You attempted to reach www.abc.com, but instead you actually reached a server identifying itself as abc.com. This may be caused by a misconfiguration on the server or by something more serious. An attacker on your network could be trying to get you to visit a fake (and potentially harmful) version of adgate.kfu.edu.sa. You should not proceed"
    so i know as this certficate is for cn=abc.com that is why we are getting such errors/status in ssl certficate.
    Now my question is
    1. Is is possible to  remove above errors doing some ssl configuration on ACE?
    2. OR we have to go for VerisgnWildcard Secure Site Pro Certificate  for CSR generated uisng cn =abc.com to be installed on ACE  and will be used  for all servers like  www.abc.com , a.abc.com etc..
    Thanks
    Waliullah

    If you want to use the same VIP and port number for multiple FQDNs, then you will need to get a wildcard certificate.  Currently, if you enter www.abc.com in your browser, that is what the browser expects to see in the certificate.  And right now it won't beause your certificate is for abc.com.  You need a wildcard cert that will be for something like *.abc.com.
    Hope this helps,
    Sean

  • Is it possible to combine multiple Pur Reqs for multiple materials to one vendor into one PO through MD07?

    We have material planners who use MD07 (traffic signal screen) Stock/Requirements List, Collective Access tab to review material requirements by MRP Controller.  Once the list is created, they select individual materials and review them in MD04.  In a lot of cases, Purchase Reqs (PRs) are reviewed and adopted into POs individually for each material.  Sometimes they will combine multiple PRs for one material into one PO.  This is a good practice.  However, from what I have found, it is not possible to combine multiple PRs for multiple materials into one PO.  If this was done it would dramatically reduce the number of POs that we are issuing to suppliers.  Problem statement:  We issue too many POs which causes additional influxes down the line (receiving dock/goods receipts processing, receiving issues, invoices, invoice issues, etc.).  Does anyone know of a way to address this problem without a major overhaul in approach or customization of the system?  Thank you in advance!   

    Hello Michael
    As far as I know, this is not possible directly from MD07, only from the MM transactions.
    The following thread suggests the following procedure to convert several PRs into one PO:
    Conversion of multiple PR into single P.O
    Drag and drop the first PR from the item overview to the shopping card on top of ME21N,
    then open the PO detail delivery schedule tab and drag  and drop all other PR to the shopping card shown  in the schedule tab
    You can use the same procedure, calling the order conversion from MD07 and then drag and drop another PRs into this PO.
    BR
    Caetano

  • Is it possible to search for multiple folders with the same name and...

    Is it possible to search for multiple folders with the same name and then select them all and change the permissions on just those folders .i.e. Search for the budget folders in all client folders and lock them down to just the project managers. Without having to go to each folder and apply the permissions.

    user11919409 wrote:
    Is it possible to create a Clone database with the same name of source db using RMAN ...
    yes
    >
    DB version is 11.2.0.2
    Is it possible to clone a 11.2.0.2 database to 11.2.0.3 home location directly on a new server . If it starts in a upgrade mode , it is ok ....yes
    Handle:     user11919409
    Status Level:     Newbie (10)
    Registered:     Dec 7, 2009
    Total Posts:     102
    Total Questions:     28 (22 unresolved)
    why do you waste time here when you rarely get any answers to your questions?

Maybe you are looking for

  • Opening/Closing a FileInputStream

    Hi all, We have a .properties file that should be checked from a servlet. The servlet will be hit by a load balancer every 5 seconds to see if the current node is active. The old programmer in the company had it like this: Properties propsFile = new

  • Can see, but can't print to network printer in Panther Server

    I'm running Panther Server 10.3.9 in a vLAN with 16 OS 9 clients. They all print to 2 Xerox Phaser 6200 and 6250 network printers. I am unable to print to either one from a G4 running Server 10.3.9. I have limited desk space so a USB-attached printer

  • Unusual black border around anything selected.

    everytime I click on something a odd black border appears. I dont know what it is and it hasnt happened before.. any clue what it is and how to get rid of it? also in the apple drop down list my restart, shutdown, and logout options appear twice.. no

  • How to display the selection screen fields for selected checkboxes

    Hi all,          I have 7 checkboxes, for each check box we have some seletion screen fields.if i select first check box,i want to display first slection screen fields only. and if we select more than one check box how to display the selection screen

  • Passing array to call library function running on VxWorks (cRIO)

    Hello, I am using a cRIO 9012 running VxWorks. I have a Call library function VI in my application and I want to pass arrays of doubles to this function. The problem is that I can not access the values in the array. Here is the source code of my func