nvidia-open-driver-G06-signed-cuda-azure-devel-570.86.15-150600.3<>,gl%p9|x HʰN0(n槀z)Y SdaA\O 2ePWg{N`, ~qzax^':gPY2[䪞bT!h,aڴq?uJ[ܩ[_rX߻%t}4ǡ{#ӺB'zRR=D  Z:]ϣ$g9^8 F9T֛CEF5qKX6x><_X?_Hd1; I }3 IW       YdK(h8p9:"G 4HIXY\4]!^>bFcGjdHeH fHlHuH$vOwP xWy^^^__DCnvidia-open-driver-G06-signed-cuda-azure-devel570.86.15150600.3.34.1Devel Package to nvidia-open-driver-G06-signed-cudaProvide build requiresments to build against nvidia-open-driver-G06-signed-cudagl%h03-ch2boSUSE Linux Enterprise 15SUSE LLC GPL-2.0-only AND MIThttps://www.suse.com/System/Kernelhttps://github.com/NVIDIA/open-gpu-kernel-modules/linuxx86_64'FM. o j&D H /nF# - D.%?&l <|lN!a=||p_1a2@w7LA Qn &&I RD?zqe=)>27sp "#|BC 8'V dN6DF? '"G; |X$!TBD???cRTY/X'#3W  <=pH&K Z F=1 a*Af ̴Q)pt2 F8>bC s 8+aFf `5u%e@wJC ( ) a XCNK=h_'5 6u(n6%N-96"NM2' Sh]53}#t?V  iFLL%") #[45`8SIW\%L$0J& X@2+[,hB+y-.O1v%e$ l`e]@e\eReDeB=e.w@e&@e&@ee d@d@d@dgd@dNdNd@dZ5dGd%yc1ccױ@c_cc@c0c0c0c0c0c@c@c@cccj@cc~ @cr-cn9@cN@cN@cAf@bc@bb֜b֜bK@b@b@blb@bbbb@b=b@b1@b1@b1@b1@b1@b1@b@b@b@b~H@b~H@b|sndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comeich@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comeich@suse.comsndirsch@suse.comsndirsch@suse.comeich@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comeich@suse.comsndirsch@suse.comeich@suse.comsndirsch@suse.comeich@suse.comsndirsch@suse.comeich@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comdimstar@opensuse.orgsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comeich@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comlnussel@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comgmbr3@opensuse.orgsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comgmbr3@opensuse.orggmbr3@opensuse.orggmbr3@opensuse.orgsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comgmbr3@opensuse.orgsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comsndirsch@suse.comtiwai@suse.comsndirsch@suse.compatrik.jakobsson@suse.comsndirsch@suse.com- fixed build against openSUSE:Factory by defining %_builddir if needed (boo#1236746)- no longer try to patch %post/%postun; no changes needed any longer - trigger script needs to be triggered now by nvidia-common-G06, which replaced kernel-firmware-nvidia-gspx-G06{,-cuda} packages- fixed version of 'Provides: nvidia-open-signed-kmp = ' of nvidia-open-signed-cuda-kmp in preamble file - fixed warning: File listed twice: /usr/src/kernel-modules/nvidia-570.86.16-default- update non-CUDA variant to 570.86.16 (boo#1236658) - supersedes aarch64-TW-buildfix.patch - supersedes set-FOP_UNSIGNED_OFFSET-for-nv_drm_fops.fop_flags.patch- Update to 565.77 - non-CUDA variant: * get rid of modproee.d and dracut.cd files and udev magic; instead require nvidia-common-G06 * Supplements: switch to really supported devices; not only the initially supported ones without graphical output - update non-CUDA and CUDA variant to 570.86.15- preamble: let -cuda KMP conflict with no-cuda variants < 550.135 (bsc#1236191)- Update to 550.144.03 (bsc#1235461, boo#1235871) * fixes CVE-2024-0131, CVE-2024-0147, CVE-2024-0149, CVE-2024-0150, CVE-2024-53869- set-FOP_UNSIGNED_OFFSET-for-nv_drm_fops.fop_flags.patch * needed for Kernel 6.12 for 565.57.01 driver; already fixed for 550.142 - -> https://forums.developer.nvidia.com/t/patch-for-565-57-01-linux-kernel-6-12/313260- Update to 550.142 (boo#1234675)- for SLE Micro 6.x use '%if 0%{?suse_version} == 1600 && !0%{?is_opensuse}'; SLE16 doesn't set %sle_version and SLE Micro 6.x is already SLE16!- Do not set %{?linux_make_arch} for running make modules/modules-install - supersedes 550.135.patch- obsolete <=560.35.03 -cuda KMPs; mistakenly we released 560.35.03 for SP4 ...- Make sure the correct FW package is installed on non-CUDA.- kmp-trigger.sh: * avoid to return with exit code != 0 if modules could not be unloaded for some reason- only obsolete 555 CUDA driver/firmware packages- preamble: no longer need to provide nvidia-open-driver-G06-kmp because since 565.57.01 to the list of requires of nvidia-compute-G06 the -signed packages has been added- For CUDA update version to 565.57.01- Add 'dummy' firmware package on SLE to work around update issues. On SLE, the firmware is installed directly from an NVIDIA-hosted repo.- preamble * resolve self conflicts of -cuda KMP during update by adding obsoletes to older versions (boo#1233332)- fixed my wrong patch 550.135.patch I introduced right below; it fixed x86_64 build more-or-less accidently but broke aarch64 build completely ...- Improve handling of conflicts between different flavors (gfx vs. CUDA) (bsc#1233332).- Update to 550.135 (boo#1233673) - 550.135.patch: * fixes wrong logic for checking supported architectures- kmp-trigger.sh: * avoid to return with exit code != 0 if no modules are loaded- Update to 550.127.05 (boo#1232057) * Fixed a bug which could cause applications using GBM to crash when running with nvidia-drm.modeset=0.- cuda-flavor provide also nvidia-open-driver-G06-kmp-$flavor = %version to workaround broken cuda-drivers- For CUDA update version to 560.35.03 - supersedes kernel-6.10.patch- cuda-flavor * provide nvidia-open-driver-G06-kmp = %version to workaround broken cuda-drivers - nv-prefer-signed-open-driver * added comments for requirements- latest change hardcoded to 555.42.06; we no longer need this for 560- nv-prefer-signed-open-driver: * added specicic versions of cuda-drivers/cuda-drivers-xxx as preconditions for requiring specific version of nvidia-compute-G06- nv-prefer-signed-open-driver: * no longer require a specific version of nvidia-open-driver-G06-signed-cuda-kmp, so it can select the correct open driver KMP matching the cuda-runtime version - cuda-flavor: * added nvidia-compute-G06 = %version to preconditions for requiring kernel-firmware-nvidia-gspx-G06, since nvidia-compute-utils-G06 does not have a version-specific requires on nvidia-compute-G06- cuda-flavor: * require kernel-firmware-nvidia-gspx-G06 instead of kernel-firmware-nvidia-gspx-G06-cuda (which provides also kernel-firmware-nvidia-gspx-G06) * trigger removal of driver modules also on kernel-firmware-nvidia-gspx-G06- no longer hard-require kernel firmware package, but install it automatically once nvidia-compute-utils-G06 gets installed - trigger removal of driver modules with non-existing or wrong firmware when (new) firmware gets installed- Update to 550.120 (boo#1230779) * Fixed a bug that could cause kernel crashes upon attempting KMS operations through DRM when nvidia_drm was loaded with modeset=0. - aarch64-TW-buildfix.patch * fixes build on aarch64 with latest TW kernel- CUDA build: removed entries from pci_ids-555.42.06 since this is doing more harm than benefit (bsc#1230368)- For CUDA (preamble file): * added: Provides: nvidia-open-driver-G06-signed-cuda-kmp-$flavor = %version which is needed for 'zypper install = ' * added: Provides/Conflicts: nvidia-open-driver-G06-signed-kmp-$flavor = %version useful for containers- reverted CUDA update version to 560.x.y due to changes in CUDA repository with CUDA 12.6/560.x.y drivers - kernel-6.10.patch: * fixes build of 555.42.06 against Kernel 6.10- For CUDA update version to 560.35.03- Update to 550.107.02 (boo#1229716)- For CUDA update version to 560.28.03- better summary and description for KMP- Update to version 555.42.06 for CUDA. This tag has become available in github, now.- Update to 550.100 (boo#1227575) * Fixed a bug that caused OpenGL triple buffering to behave like double buffering.- Use macro which is set during build but not when running `osc service run download_files` to make sure both source versions are visible for download. This way, only the correct source tarball will be included in the source package of its respective flavor but both can be downloaded (updated) simultaniously.- make buildservice happy; all sources need to be mentioned as such in specfile- To avoid issues with missing dependencies when no CUDA repo is present make the dependecy to nvidia-compute-G06 conditional. - CUDA is not available for Tumbleweed, exclude the build of the cuda flavor.- preamble: let the -cuda flavor KMP require the -cuda flavor firmware- Add a second flavor for building the kernel module versions used by CUDA. The kmp targetting CUDA contains '-cuda' in its name to track its versions separately from the graphics kmp. (bsc#1227417) - Provide the meta package nv-prefer-signed-open-driver to make sure the latest available SUSE-build open driver is installed - independent of the latest available open driver version in he CUDA repository. Rationale: The package cuda-runtime provides the link between CUDA and the kernel driver version through a Requires: cuda-drivers >= %version This implies that a CUDA version will run withany kernel driver version equal or higher than a base version. nvidia-compute-G06 provides the glue layer between CUDA and a specific version of he kernel driver both by providing a set of base libraries and by requiring a specific kernel version. 'cuda-drivers' (provided by nvidia-compute-utils-G06) requires an unversioned nvidia-compute-G06. With this, the resolver will install the latest available and applicable nvidia-compute-G06. nv-prefer-signed-open-driver then represents the latest available open driver version and restricts the nvidia-compute-G06 version to it. (bsc#1227419)- Security Update 550.90.07 (boo#1223356) [CVE-2024-0090, CVE-2024-0091, CVE-2024-0092]- Update to 550.78 * addresses boo#1223454- Update to 550.76 * addresses boo#1222972- Update to 550.67 - rename pci_ids-unsupported* to pci_ids-supported* files- let's provide/obsolete nvidia-open-driver-G06 instead of nvidia-open-driver-G06-kmp since older versions still had conflicts to nvidia-open-driver-G06-kmp ...- Update to 550.54.14 * Added vGPU Host and vGPU Guest support. For vGPU Host, please refer to the README.vgpu packaged in the vGPU Host Package for more details. - supersedes the following patches: * 0001-Don-t-override-INSTALL_MOD_DIR.patch * kernel-6.7.patch- re-enable build of -azure kernel flavor; syntax check was wrong- remove conflicts to nvidia-open-driver-G06-kmp, since it's now provided instead (OMG!); add obsoletes to it as well to make sure it gets replaced (bsc#1220196)- Use %autosetup macro. Allows to eliminate the usage of deprecated %patchN- kernel-syms-azure is not available on ALP- provide nvidia-open-driver-G06-kmp = %version (jsc#PED-7117) * this makes it easy to replace the package from nVidia's CUDA repository with this presigned package- splitted up 61-nvidia-$flavor.conf to 59-nvidia-$flavor.conf and 61-nvidia-$flavor.conf, because 'install' line cannot be overwritten with higher config number ...- mistakenly moved dracut config file from 60-nvidia-%1.conf to 61-nvidia-%1.conf --> reverted!- switched from 60-nvidia-$flavor.conf to 61-nvidia-$flavor.conf in modprobe.d to resolve conflict with older package, which can be installed in parallel- kernel-6.7.patch * fixes build against kernel 6.7 (boo#1219117)- create /run/udev/static_node-tags/uaccess/nvidia${devid} symlinks also during modprobing the nvidia module; this changes the issue of not having access to /dev/nvidia${devid}, when gfxcard has been replaced by a different gfx card after installing the driver- fix typo in changelog: * NVreg_OpenRMEnableSupporteGpus -> NVreg_OpenRmEnableUnsupportedGpus (boo#1215981, comment#33)- Update to 545.29.06- no longer try to overwrite NVreg_OpenRmEnableUnsupportedGpus driver option setting; apparently it's ignored by the driver (boo#1215981, comment#26)- use different modprobe.d config file to resolve conflict with older driver package (boo#1217370); overwrite NVreg_OpenRmEnableUnsupportedGpus driver option setting (disable it), since letting it enabled is supposed to break booting (boo#1215981, comment#23)- Update to 545.29.02 - added fbdev=1 option for nvidia-drm module, which gives us a proper framebuffer console now ... - nosimplefb kernel option no longer needed with usage of nvidia-drm's fbdev=1 option - nvidia's NVreg_OpenRmEnableUnsupportedGpus=1 option no longer needed; GeForce and Workstation GPUs now officially supported - support added for H100/H800 GPUs (Hopper)- cleanup; sle12 no longer supported- Update to version 535.129.03- cleanup: get rid of possibiliy to hardcode pci list; it's no longer required anyway; secureboot signing has been fixed ... - removed pci-table.patch- Add a devel package so other modules can be built against this one. [jira#PED-4964]- disabled build of nvidia-peermem module; it's no longer needed and never worked anyway (it was only a stub) [boo#1211892]- preamble: added conflict to nvidia-gfxG05-kmp to prevent users from accidently installing conflicting proprietary kernelspace drivers from CUDA repository- Update to version 535.113.01- kmp-post.sh/kmp-postun.sh: * add/remove nosimplefb=1 kernel option in order to fix Linux console also on sle15-sp6/Leap 15.6 kernel, which will come with simpledrm support- Update to version 535.104.05- require one-time renamed package kernel-firmware-nvidia-gspx-G06 from now on (boo#1212724)- Replace transitional %usrmerged macro with regular version check (boo#1206798)- Update to version 535.86.05- _constraints: 4GB wasn't enough; trying 8GB now ...- _constraints: reserve 4GB of disk space for building in the hope that this will be also sufficient when building for sle15-sp5- Update to version 535.54.03- Update to version 525.116.04- Update to version 525.116.03- Update to version 525.105.17- no longer use %hardcode_pci_list macro on sle15/Leap15, since the issue with generating wrong PCI ID modaliases for Supplements has been fixed meanwhile also in pesign-obs-integration package for sle15/Leap15 (boo#1207520)- Update to version 525.89.02- no longer use %hardcode_pci_list macro on TW, since the issue with generating wrong PCI ID modaliases for Supplements has been fixed now in pesign-obs-integration package for TW (boo#1207520)- introduce %hardcode_pci_list macro for hardcoding PCI ID modaliases for Supplements; make it default to *temporarily* workaround boo#1207520 * TW: enable only officially supported headless GPUs! * Leap/sle: enable all Turing and Ampere GPUs and set the appropriate modprobe.d option, so we can at least begin testing on sle15-sp5 ... - adds pci-table.patch, generati-pci-table.sh- added pesign-copy-sources, pesign-spec-macros, which are needed to fix repackaging step required for securebooot signing kernel modules (boo#1207520) - pesign-spec-macros: used for defining RPM macros; here: %define __kmp_supplements %_sourcedir/my-find-supplements \ %_sourcedir/pci_ids-%{version} - pesign-copy-sources: sources needed by the RPM macros above; here: pci_ids-%{version}, my-find-supplements - all above needs changes for pesign-obs-integration first (SR#1060443)- added conflicts to nvidia-driver-G06-kmp package (boo#1207495)- Update to version 525.85.05 * Fix build problems with Clang 15.0, #377 by @ptr1337- require "group(video)"- set options for nvidia kernel module via modprobe install script; no longer adjusting of nvidia's NVreg_DeviceFileGID option in modprobe.d file needed in %post- package the modprobe.d file as a sample in /usr/lib/nvidia-open-driver-G06-signed; then adjust GUID of video group for nvidia's NVreg_DeviceFileGID option and write the result to /etc/modprobe.d, i.e. no longer touch a packaged modprobe.d file below /usr- set group id for nvidia's NVreg_DeviceFileGID option already during build, since it's not allowed to edit modprobe.d files during %post- added pci_ids-unsupported-525.78.01/pci_ids-unsupported as sources to specfile- added empty %changelog section to specfile- implements jsc#PED-2658- require kernel-firmware-nvidia-gsp-G06 (with suffix -G06!)- Update to version 525.78.01 * adds alpha support for + NVIDIA GeForce RTX 3050 + NVIDIA GeForce RTX 4090 + NVIDIA RTX 6000 Ada Generation + NVIDIA L40 + NVIDIA GeForce RTX 4080 + NVIDIA GeForce RTX 4070 Ti- renamed package to nvidia-open-driver-G06-signed due to package redesign- Update to version 525.60.11- set NVreg_PreserveVideoMemoryAllocations kernel option now needed for GNOME Wayland (gdm) since commit 51181871e9db716546e9593216220389de0d8b03 Author: Ray Strode Date: Fri Mar 4 14:11:03 2022 -0500 data: Disable wayland on nvidia if suspend is broken- added /etc/dracut.conf.d/60-nvidia-$flavor.conf in order to omit nvidia modules in initrd, which apparently were added automatically - apart from nvidia-uvm module; related to boo#1173733- Update to 525.53 * GSP firmware is now distributed as multiple firmware files: this release has gsp_tu10x.bin and gsp_ad10x.bin replacing gsp.bin from previous releases. + Each file is named after a GPU architecture and supports GPUs from one or more architectures. This allows GSP firmware to better leverage each architecture's capabilities. + The .run installer will continue to install firmware to /lib/firmware/nvidia/ and the nvidia.ko kernel module will load the appropriate firmware for each GPU at runtime. * Add support for IBT (indirect branch tracking) on supported platforms, #256 by @rnd-ash * Return EINVAL when [failing to] allocating memory, [#280] by @YusufKhan-gamedev * Fix various typos in nvidia/src/kernel, #16 by @alexisgeoffrey * Added support for rotation in X11, Quadro Sync, Stereo, and YUV 4:2:0 on Turing.- (build-)require perl-Bootloader due to pbl usage in %post(-un)- kmp-post.sh/kmp-postun.sh: * add/remove nosimplefb=1 kernel option in order to fix Linux console on TW kernel, which comes with simpledrm support- Update to 515.76 * Improved compatibility with new Linux kernel releases * Fixed possible excessive GPU power draw on an idle X11 or Wayland desktop when driving high resolutions or refresh rates- added pci_ids-unsupported -> pci_ids-unsupported-515.65.01 symlink to which I can refer from blog- Update to 515.65.01 * Collection of minor fixes to issues, #6 by @Joshua-Ashton * Remove unnecessary use of acpi_bus_get_device().- Force CC to gcc to avoid compiler differs warnings- json-to-pci-id-list.py * added new Python script by Tamara Schmitz which is being used now to parse supported-gpus.json in order to create the PCI ID list 'pci_ids-unsupported-'- pci_ids-unsupported-515.57 * currently unsupported devices (Turing and later, which needs NVreg_OpenRmEnableUnsupportedGpus=1 kernel option; Alpha Quality!)- define %kernel_module_directory macro on older distros in order to fix build on Leap 15.3(GA)/sle 15 SP3(GA)- update to 515.57 * Backtick is deprecated, #273 by @arch-user-france1- Revert last change and use default CA [pesign-obs-integration and kernel-source docs are incorrect]- Create uefi cert package- Sign the kernel modules and compress them- don't buildrequire azure on aarch64; no such kernel flavor available on aarch64- move cuda-drivers = %version provide to nvidia-computeG0X, which then again requires nvidia-gfxG0X-kmp or nvidia-open-gfxG0X-kmp respectively- preamble: provide cuda-drivers = %version, so you can just install 'cuda' meta package to install CUDA stack (cuda --> cuda-11-7 --> cuda-runtime-11-7 --> cuda-drivers >= 515.43.04) from https://developer.download.nvidia.com/compute/cuda/repos/opensuse15/x86_64/- preamble: added Conflicts to nvidia-gfxG06-kmp- update to 515.48.07 * List of compatible GPUs in README.md. * Fix various README capitalizations, #8 by @lx-is * Automatically tag bug report issues, #15 by @thebeanogamer * Improve conftest.sh Script, #37 by @Nitepone * Update HTTP link to HTTPS, #101 by @alcaparra * moved array sanity check to before the array access, #117 by @RealAstolfo * Fixed some typos, #122 by @FEDOyt * Fixed capitalization, #123 by @keroeslux * Fix typos in NVDEC Engine Descriptor, #126 from @TrickyDmitriy * Extranous apostrohpes in a makefile script [sic], #14 by @kiroma * HDMI no audio @ 4K above 60Hz, #75 by @adolfotregosa * dp_configcaps.cpp:405: array index sanity check in wrong place?, #110 by @dcb314 * NVRM kgspInitRm_IMPL: missing NVDEC0 engine, cannot initialize GSP-RM, #116 by @kfazz * ERROR: modpost: "backlight_device_register" [...nvidia-modeset.ko] undefined, #135 by @sndirsch * aarch64 build fails, #151 by @frezbo - supersedes backlight-workaround.patch- Set mno-outline-atomics to fix libgcc issues on aarch64- preamble: require kernel-firmware-nvidia-gsp package for nvidia's versioned "gsp.bin" fimrware- enable build also for aarch64- improved summary and description- modprobe.nvidia.install * install script when the "nvidia" module is being loaded; needed for creating devices - specfile * filled modprobe.d config file (options and install script)- kmp-post.sh/kmp-postun.sh * additional code for %post/%postun scriptlets to create nvidia devices - specfile: adjust %kernel_module_package macro appropriately- added kmp-filelist - adding an empty modprobe.d config file (to be filled later)- added hardware supplements * added my-find-supplements, pci_ids-515.43.04 * changes in specfile to generate the hardware supplements from pci_ids-515.43.04 via my-find-supplements- Set SYSSRC and SYSOUT for building with proper kernel source tree for each flavor - Exclusive build for x86_64 - backlight-workaround.patch: workaround for a build failure with azure flavor - persistent-nvidia-id-string.patch: make persistent nvidia id string for reproducible builds- switched to a known license to fix build: MIT/GPLv2 --> GPL-2.0 and MIT- 0001-Don-t-override-INSTALL_MOD_DIR.patch * Make sure INSTALL_MOD_DIR from our spec-file is honoured - Make the default flavor build. Not sure how to get other flavors to build.- created package * build still fails completely * more things are missing  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~570.86.15-150600.3.34.1570.86.15-150600.3.34.1       kernel-modulesnvidia-570.86.15-azureKbuildMakefilecommonincconftest.hcpuopsys.hnv-caps.hnv-chardev-numbers.hnv-dmabuf.hnv-firmware-registry.hnv-firmware.hnv-gpu-info.hnv-hash.hnv-hypervisor.hnv-ioctl-numa.hnv-ioctl-numbers.hnv-ioctl.hnv-kernel-interface-api.hnv-kref.hnv-kthread-q-os.hnv-kthread-q.hnv-linux.hnv-list-helpers.hnv-lock.hnv-memdbg.hnv-mm.hnv-modeset-interface.hnv-msi.hnv-pci-types.hnv-pci.hnv-pgprot.hnv-procfs-utils.hnv-procfs.hnv-proto.hnv-retpoline.hnv-time.hnv-timer.hnv.hnvCpuUuid.hnv_speculation_barrier.hnv_stdarg.hnv_uvm_interface.hnv_uvm_types.hnvgputypes.hnvkms-api-types.hnvkms-format.hnvkms-kapi.hnvlimits.hnvmisc.hnvstatus.hnvstatuscodes.hnvtypes.hosos-interface.hnv_memory_area.hnv_memory_type.hrm-gpu-ops.hrs_access.hconftest.shcount-lines.mkdkms.confheader-presence-tests.mknvidianvidia-drmnv-kthread-q.cnv-pci-table.cnv-pci-table.hnv_common_utils.hnvidia-dma-fence-helper.hnvidia-dma-resv-helper.hnvidia-drm-conftest.hnvidia-drm-connector.cnvidia-drm-connector.hnvidia-drm-crtc.cnvidia-drm-crtc.hnvidia-drm-drv.cnvidia-drm-drv.hnvidia-drm-encoder.cnvidia-drm-encoder.hnvidia-drm-fb.cnvidia-drm-fb.hnvidia-drm-fence.cnvidia-drm-fence.hnvidia-drm-format.cnvidia-drm-format.hnvidia-drm-gem-dma-buf.cnvidia-drm-gem-dma-buf.hnvidia-drm-gem-nvkms-memory.cnvidia-drm-gem-nvkms-memory.hnvidia-drm-gem-user-memory.cnvidia-drm-gem-user-memory.hnvidia-drm-gem.cnvidia-drm-gem.hnvidia-drm-helper.cnvidia-drm-helper.hnvidia-drm-ioctl.hnvidia-drm-linux.cnvidia-drm-modeset.cnvidia-drm-modeset.hnvidia-drm-os-interface.cnvidia-drm-os-interface.hnvidia-drm-priv.hnvidia-drm-sources.mknvidia-drm-utils.cnvidia-drm-utils.hnvidia-drm.Kbuildnvidia-drm.cnvidia-drm.hnvidia-modesetnv-kthread-q.cnvidia-modeset-linux.cnvidia-modeset-os-interface.hnvidia-modeset.Kbuildnvkms-ioctl.hnvkms.hnvidia-peermemnv-p2p.hnvidia-peermem.Kbuildnvidia-peermem.cpeer_mem.hnvidia-uvmcla06fsubch.hcla16f.hclb069.hclb06f.hclb0b5.hclc06f.hclc076.hclc0b5.hclc1b5.hclc365.hclc369.hclc36f.hclc3b5.hclc46f.hclc56f.hclc5b5.hclc6b5.hclc7b5.hclc86f.hclc8b5.hclc96f.hclc9b5.hclca6f.hclcab5.hclcba2.hctrl2080mc.hhwrefamperega100dev_fault.hdev_runlist.hblackwellgb100dev_fault.hhoppergh100dev_fault.hdev_mmu.hmaxwellgm107dev_mmu.hpascalgp100dev_fault.hdev_fb.hdev_mmu.hturingtu102dev_fault.hdev_mmu.hvoltagv100dev_fault.hdev_fb.hdev_mmu.hnv-kthread-q-selftest.cnv-kthread-q.cnvCpuUuid.cnvidia-uvm-sources.Kbuildnvidia-uvm.Kbuildnvstatus.cuvm.cuvm.huvm_ada.cuvm_ada_fault_buffer.huvm_ampere.cuvm_ampere_ce.cuvm_ampere_fault_buffer.cuvm_ampere_fault_buffer.huvm_ampere_host.cuvm_ampere_mmu.cuvm_api.huvm_ats.cuvm_ats.huvm_ats_faults.cuvm_ats_faults.huvm_ats_ibm.cuvm_ats_ibm.huvm_ats_sva.cuvm_ats_sva.huvm_blackwell.cuvm_blackwell_fault_buffer.cuvm_blackwell_fault_buffer.huvm_blackwell_host.cuvm_blackwell_mmu.cuvm_ce_test.cuvm_channel.cuvm_channel.huvm_channel_test.cuvm_common.cuvm_common.huvm_conf_computing.cuvm_conf_computing.huvm_extern_decl.huvm_fault_buffer_flush_test.cuvm_forward_decl.huvm_get_rm_ptes_test.cuvm_global.cuvm_global.huvm_gpu.cuvm_gpu.huvm_gpu_access_counters.cuvm_gpu_access_counters.huvm_gpu_isr.cuvm_gpu_isr.huvm_gpu_non_replayable_faults.cuvm_gpu_non_replayable_faults.huvm_gpu_replayable_faults.cuvm_gpu_replayable_faults.huvm_gpu_semaphore.cuvm_gpu_semaphore.huvm_gpu_semaphore_test.cuvm_hal.cuvm_hal.huvm_hal_types.huvm_hmm.cuvm_hmm.huvm_hopper.cuvm_hopper_ce.cuvm_hopper_fault_buffer.cuvm_hopper_fault_buffer.huvm_hopper_host.cuvm_hopper_mmu.cuvm_hopper_sec2.cuvm_host_test.cuvm_ioctl.huvm_kvmalloc.cuvm_kvmalloc.huvm_kvmalloc_test.cuvm_linux.cuvm_linux.huvm_linux_ioctl.huvm_lock.cuvm_lock.huvm_lock_test.cuvm_map_external.cuvm_map_external.huvm_maxwell.cuvm_maxwell_access_counter_buffer.cuvm_maxwell_ce.cuvm_maxwell_fault_buffer.cuvm_maxwell_host.cuvm_maxwell_mmu.cuvm_maxwell_sec2.cuvm_mem.cuvm_mem.huvm_mem_test.cuvm_migrate.cuvm_migrate.huvm_migrate_pageable.cuvm_migrate_pageable.huvm_mmu.cuvm_mmu.huvm_page_tree_test.cuvm_pascal.cuvm_pascal_ce.cuvm_pascal_fault_buffer.cuvm_pascal_fault_buffer.huvm_pascal_host.cuvm_pascal_mmu.cuvm_peer_identity_mappings_test.cuvm_perf_events.cuvm_perf_events.huvm_perf_events_test.cuvm_perf_heuristics.cuvm_perf_heuristics.huvm_perf_module.cuvm_perf_module.huvm_perf_module_test.cuvm_perf_prefetch.cuvm_perf_prefetch.huvm_perf_thrashing.cuvm_perf_thrashing.huvm_perf_utils.cuvm_perf_utils.huvm_perf_utils_test.cuvm_pmm_gpu.cuvm_pmm_gpu.huvm_pmm_sysmem.cuvm_pmm_sysmem.huvm_pmm_sysmem_test.cuvm_pmm_test.cuvm_policy.cuvm_populate_pageable.cuvm_populate_pageable.huvm_processors.cuvm_processors.huvm_procfs.cuvm_procfs.huvm_pte_batch.cuvm_pte_batch.huvm_push.cuvm_push.huvm_push_macros.huvm_push_test.cuvm_pushbuffer.cuvm_pushbuffer.huvm_range_allocator.cuvm_range_allocator.huvm_range_allocator_test.cuvm_range_group.cuvm_range_group.huvm_range_group_tree_test.cuvm_range_tree.cuvm_range_tree.huvm_range_tree_test.cuvm_rb_tree.cuvm_rb_tree.huvm_rb_tree_test.cuvm_rm_mem.cuvm_rm_mem.huvm_rm_mem_test.cuvm_sec2_test.cuvm_test.cuvm_test.huvm_test_ioctl.huvm_test_rng.cuvm_test_rng.huvm_thread_context.cuvm_thread_context.huvm_thread_context_test.cuvm_tlb_batch.cuvm_tlb_batch.huvm_tools.cuvm_tools.huvm_tools_init.huvm_tracker.cuvm_tracker.huvm_tracker_test.cuvm_turing.cuvm_turing_access_counter_buffer.cuvm_turing_fault_buffer.cuvm_turing_fault_buffer.huvm_turing_host.cuvm_turing_mmu.cuvm_types.huvm_unit_test.huvm_user_channel.cuvm_user_channel.huvm_va_block.cuvm_va_block.huvm_va_block_test.cuvm_va_block_types.huvm_va_policy.cuvm_va_policy.huvm_va_range.cuvm_va_range.huvm_va_range_device_p2p.cuvm_va_space.cuvm_va_space.huvm_va_space_mm.cuvm_va_space_mm.huvm_volta.cuvm_volta_access_counter_buffer.cuvm_volta_ce.cuvm_volta_fault_buffer.cuvm_volta_fault_buffer.huvm_volta_host.cuvm_volta_mmu.cdetect-self-hosted.hexport_nvswitch.hhallibrarycryptlibcryptlib_aead.hcryptlib_cert.hcryptlib_dh.hcryptlib_ec.hcryptlib_ecd.hcryptlib_hash.hcryptlib_hkdf.hcryptlib_mac.hcryptlib_rng.hcryptlib_rsa.hcryptlib_sm2.hi2c_nvswitch.cinternallibspdm_lib_config.hinternal_crypt_lib.hioctl_common_nvswitch.hioctl_nvswitch.hlibrarycryptlib.hspdm_lib_config.hlibspdm_aead.clibspdm_aead_aes_gcm.clibspdm_ec.clibspdm_ecc.clibspdm_hkdf.clibspdm_hkdf_sha.clibspdm_hmac_sha.clibspdm_internal_crypt_lib.clibspdm_rand.clibspdm_rsa.clibspdm_rsa_ext.clibspdm_sha.clibspdm_shash.clibspdm_x509.clinux_nvswitch.clinux_nvswitch.hnv-acpi.cnv-caps-imex.cnv-caps-imex.hnv-caps.cnv-cray.cnv-dma.cnv-dmabuf.cnv-host1x.cnv-i2c.cnv-kthread-q.cnv-memdbg.cnv-mmap.cnv-modeset-interface.cnv-msi.cnv-nano-timer.cnv-p2p.cnv-p2p.hnv-pat.cnv-pat.hnv-pci-table.cnv-pci-table.hnv-pci.cnv-procfs.cnv-reg.hnv-report-err.cnv-report-err.hnv-rsync.cnv-rsync.hnv-usermap.cnv-vm.cnv-vtophys.cnv.cnv_gpu_ops.hnv_uvm_interface.cnvidia-sources.Kbuildnvidia.Kbuildnvlink_caps.cnvlink_caps.hnvlink_common.hnvlink_errors.hnvlink_export.hnvlink_linux.cnvlink_linux.hnvlink_os.hnvlink_pci.hnvlink_proto.hnvspdm_cryptlib_extensions.hos-interface.cos-mlock.cos-pci.cos-registry.cos-usermap.cprocfs_nvswitch.crmp2pdefines.h/usr/src//usr/src/kernel-modules//usr/src/kernel-modules/nvidia-570.86.15-azure//usr/src/kernel-modules/nvidia-570.86.15-azure/common//usr/src/kernel-modules/nvidia-570.86.15-azure/common/inc//usr/src/kernel-modules/nvidia-570.86.15-azure/common/inc/os//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-drm//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-modeset//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-peermem//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/ampere//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/ampere/ga100//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/blackwell//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/blackwell/gb100//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/hopper//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/hopper/gh100//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/maxwell//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/maxwell/gm107//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/pascal//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/pascal/gp100//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/turing//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/turing/tu102//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/volta//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia-uvm/hwref/volta/gv100//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia/hal//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia/hal/library//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia/hal/library/cryptlib//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia/internal//usr/src/kernel-modules/nvidia-570.86.15-azure/nvidia/library/-fmessage-length=0 -grecord-gcc-switches -O2 -Wall -D_FORTIFY_SOURCE=2 -fstack-protector-strong -funwind-tables -fasynchronous-unwind-tables -fstack-clash-protection -gobs://build.suse.de/SUSE:Maintenance:37300/SUSE_SLE-15-SP6_Update/0e5a1c21886a5bc9b6ad54a9b9f7f212-nvidia-open-driver-G06-signed.SUSE_SLE-15-SP6_Update:cudadrpmxz5x86_64-suse-linuxdirectoryASCII textC source, ASCII textC source, UTF-8 Unicode textPOSIX shell script, ASCII text executableRutf-8328157de909e763e68aa5bd94fa8aa1be80a6a712ccc673608444a13e52e1cfc?@7zXZ !t/h]"k&48u<#}ҷgXНfj ܑ)p220XeSG ymN8U,a?>flTG֙jƒ#mS ˴Su_g9#JrTFfj IK`$}6a.%ƂQ 'J vm GLJqe5fMO..[!葾m]Jg6muE0'*˫&OsB Ej1ETpý5X QxT ̻ {@s ofv2VoEA M F3Βɷ#B b "(jR1D_fk;_>(غmzJ>7=W`K f3i|`0`7YHreYz)ul ມmTn!4жJqn!`4"rпs=`j)&=v0~ ALYF7u\X5=GscuuqEQIm;>Wu Is*o>\qؽ{wR y怋)v9 hhCV CQ*h]hKF]&8$ZؙDԕM~ ogJ0EZN")OM\xi_Е!9NzD|Kޛˠq^tSV;b_͈A@XT<[ݮOxۥX=sQҴOOad(ط#f6G&xSkn;c%P2 X.p=yAM<{B8QCO&0>%BtLq1h@CZMP%/G3Y :ssNJ2.z;Q0 [45aJZKkTa#GOEvV7r4(>fu}sW5]X_CJݕf`^H[ ed vWO}$ xwʓ1`s MҘe:ڷtº3E%ELÓ ®݀=KwQdch`w<67e͝Ec?߁;ܳ8DEx:eoqU0զ0?\[2TN6-~q'߆2Ӡ60֤U?.фob:1pb/Tm]iJehi\}x^"dN|f=C;hwT׻ fGI/*]D-Z_eweɭjאME^T͢BH#J/ PktWrs'3 17**A:p~tb93k$&fJy'}H l41?|{(JOK{T^6}"P'o$~6q dKW Rr~o?{󻷚zB1n(#)gbqB% H'u@GȚg U<Ù{:JB ݂x/\? +C)RM s k77y,,QI#|h $&3[b+.C WLc-|4(eOUM^@E5A  Al()*hܶ|U[--ݻ=8۽^91nb &Rml(^mu,1иFx?zX{`">~= |G Bhae M񨁮J2cN2ZKBf>i_B])c߰^;~d[,$i)uOzu o5^ IB6)_Jl9 l1-߳YGlH, ` #~԰Zb8M/GcOJa(k8d`W<ۃ1*sRi,žO$/sB`: mi/^  ̳՘wb-}&]@ q)xm҄_ܯ4_EAwF$1h[YeirWAsء2?rRHAK.&[;th% Y#X: éHJfSkN4!-oIG>(KV @WǕ]GEBouȼ6;۳"uG%w._ܽr&R ޕ{7`FKQPБscz’kUd6m#d*/ ; ưYjWMk^x"z&&LY蝘CpNL=<^UR-a);;V^S䵨]ݛeI 3[aزpJa`i,V]EI`Y- Ӊ|y d`I~>*>ao~X`C'yzHL8 (]UMЈw㪠bZ=զާUS^wtaW@1/5` [g|-(u;2r-,_(7=*NZ p6O4qPTD/.-5;+*G % LEv뺑h4&uJL]^4a=}o}Ex^zI0S4qx9S`nnf򕏣\@lJX~.l]Fr5U Rm.C77z   YKFĦ,#N"_cTo$ {KN]KޯGPWtA؏MғF? pۖ^$+Okjwԡ(VC:d(=:Dʼj-,ؗ.5.Ǔv|:{X dཊ 6y #{9)[9\viܒPI ?B3i-nd+l,O52@q ;-W *j)7E(G@K~I=IB3Kδ5]>4qbܜ`aU˕]c4 V?^o .]jU 0x^zb#X-qb>pv => q߫\SVfL׋ P7`̬SUC }+g=9CR$T(M2ᦪՍ_ Pw՘* ]4^o1k6=4O%[/>q@7pU& wY(U8zD|1415jT DQpXWKTw^{*l?rߠڍfORtm|t-蜆5)&,ţ Z+'0KJOq@YJl4lUjHdLb0rp""g;ji"|59h`gjr@wvjMr 6AòĬg'17ݼhtON:_:qݲ燰r} UG&9?QX|;dt]<>ybQo{]AI` 9qwI3q:ٞNeUƃ@DhU +mڇ/ν }9>@G9xM+ :#"$dWgSxPbs VTݧ1O3P)mm[ѧz}iխaQj$M_jw/Bw§c)~=n.؇1V$fFzјË;޵v f)mQNfR4dSI!@|'nqMjR&Wpϣ72UiG=‚5҂1ۉjfZP6K7a~e_8TsqRZa[AC=2&G,w)wѰJgµ%bC^h&d͏9 ,ބ 鼁Wks2 Zyܞ%sP8DLJ>6͟jEXy= ϴwND ې}3 tV!Ԝr]ꍓg 05$k`5SG5<`Y`+jou& Z_S/y~ಀ8Pʟt-(: Utp$T7AUNvZw-)' o+0uL١o|`:jo@j+F!^pQ|Ƞd3gYuܝ1~shBJd±γ:0_>ȮEeb3CDbo#8`|wkm6HErTeMdoƳd27kaIoРQ7~ >|Ԉwր;be'1%cYw{f4OkQ[$,ƓKPWfuIrVB5:EㄕN'AEM2H[gH(k!uߏ| -ӛcO!P:x<3G)юfV[$&mJWᚇP̍oJŮՒ'h0CxZ*:P2M8*6ڠv8NC-P1"(j{yk.r)=aO`fܮ3!QE~"u1 rεDKeSZ)IO&1E8o8v)eOT+.{l:*Oyu.;V<7!0nU_`gZaP&G'<B(91.kac"!poH=dX& ܱgb&wIEzc#|,GL7>5k.V!Áԯ2(Fn$}"w{FƗlc&ж)&*l8Òy~%acNE"hZ,%'ƴ Y&Mpc 9~^&) ]:F6O^n0?c_I.!nYal;D:~4 M]P;rvWs$Ijn VP$kQ6m]xRq}N\}*Ϋouy6_}N访 F]!Rp7bCA'o?.{v} yvjp^˖^Y焎0QՐi.]ׅlʴ.t`x~KyܬDoԐ3;N>1Dz{]獗 3 O%m4FߗB1̕LCL$`ZƱ XjV1n(cJ s>}'T)f$JMuZ5|h7 YeNer;cji*fݟ'kN2P'BVZ_rp#"kW^\rzڗpoDz&?O.K +`#!)Ȫd$Xb«Gra;M5kќV b^fq`Dzr=+O><'Υ 6N +k$CkMˈnv͞f=,ru@9Ut[8B q_Nxh^"Qcqk>h2z^ҫ@a!@lU{oN<s @Uxe1pzs/V"EcdXF7_'ucs. v|^qŅpzMNKk$yMc2{(3CIyvHؚ:vXC$Qiɴ"]@!{^. Yh R/= vjtC{rNñf.d@_38}U Qoh\Oge:%Jοq&Nƒv*SV4Rʕ,:-j-g6\|Iǫ\^0) ֠΂zG{Rke;l6 (<-vBRʈ FW9a#-H{,h=kF[DVsJ8<B7ʮhW&tzx YҘT\n]pǒ;uJ#m6E#$cKx3nQtQ"60]P  eIܔZJtVH!#O6Jy 6IvrK-ơEV](i+dtPe P`f{/CzGqvj2 CMrBl:G*ov O-:QY1jN@x|}]BC b!V#x @g$Z@^ht0[NUYv'Wf0ғ":erC^?N#&1 \#Fm'lE=_4lkt֬댐k~Yړ@_\9eAdOG K_a D! ϒQRؚ_sbX /b7UI ף>Od75;V+pvlc=+S wrFmv=7!,I@v jXm­Lӵ.8QV?Ű{p[}Vml(xPVS5Rzi?~0E|t{ ja("*pI$$,t;On"+8mf0Yf0tWRźbW cۉí[4@fo+( ajƘf*C"68VA棸ϾF2!C:2#|)!*Q-tZi >qDwܢGc$Jeow4>7&'u1Z&Jk޸C}_RM `R>)JU3'C ຏ?Rdȏ57n:Kn@_wisӯ ZSyqkJ= }ˆk}v`2žfbAy/7yZiI^EY]Ww6DkvV}:r|QZW)8\'?-]Px=Iu(|cfdj݅πr-%])rt!\kxn6LX߮GOJO~gvrdR} Óh1 {lf/ 섒hpU:`+i+zQ%QP.*9/*LB] 0Rʞ ʫ׺ڹIH&|aKЭ+ .'$QrN_2P誒/hG~0j5m__UY[SaCo@ve.# QeܪUa-lSfG>*h|2Hɰֶ2Fɬ!g'M֔LU/r 3ouR"i.p_sҹd(Z틀cLXZ'zjٷێ"nq"6eLY=ϥ^#yaTU:)ϽͿqظ! O*‡UkGu0/DP)O+!D ?C3p^!K!>< "7*:RĞxRKlH-Ȁ_ I3n7ki_)G* x^}(dʰ6T>mqGQӰւ$uuJr; I uŬއkF.OdRZ 9s=`BE|+U/>1S@L[ǃ[hdS3 X:XqAXyqj{fEMG鳺7@׆wv Ex"hKdf,9D'1%7=nBLRKB.Ns>m%+kPކnD{7 ,('P0cgd]}l'V\+T-J@~""}.|Qb ΌLgYӡ}ƘjD+n\yQ`>-:jFy/ʋt$b$ZQcG+>`PܑQ;$KIӐk[QɆ%3LQ\׫Z dA+@^ U0HtɪkMi;KFnK6[iJLO%b`(´/#'%Y;MV*diIDs'%Nz9%`NC{`ier3SȂ+DZתJһ|Po蚍K8LpcP"Tz-* |[ٺvc\~7Ռ"bb.c~3&Je<NꍖZx*7Ʈ +Ej$-EUU7<5:p>X隷D ~H.J̰" Y`@DOa0O5)}V/%a@?'^saC^E` [f? :H4Dyyw7u`J3 ? 5jsj3]M}]Xb.*jK:0U\T.`߃.:kn얡_MlLxfz #2GN%.Lku6Yt?Ae$bliuJtLZW2_|wzFY~32|EV%,Y\+ =w:w5({e^Gx(k8[#u8&3N^tRf@Ҿ%xB Bp _'ˮ{7޽CW3wp{?><p`7DȌ0w1 '1Y 0\]5??Dhc؟6gG.] kH}3 J3@~s+^{l #{O8zЦHզ vgkuˊ@%9mx|{%Da蓇3ED :{>"w.IMZViګk_~m>1P|(cN͇JKk0'\U^mW]ωމCSXpۑ^~ͳE"FJs&aS/%RB{02Or$PA48>:aeZ[5u`c,:u,wb8r<#D-lB|;lo y F)ҫjP_. L׽<2#l@XT{N%}+v q;^ĪKKɂqPDHJXܿ˜>UȉLu6c~qChg>s^Uthۇ OtGTK}Tc~M-sL쁶Wr9:u)fAh4&n}]'m-:ffuӾ 1 ĭܼ#gmKy{9)N%:E07@)fhs/VєvFt(!a?^"4{= RQL“^VNg׹ ?eHcGm(QvaFaDXȾ% >A¥pep~[FSУZϳT2[gX.X39.`ij#DO6ڟϺH B"{,6YLPe6:?.Q(F#+x#_YCc֋IpT&M5BASAc8U2kPݲ$ _<^B.~ZOۮVy;Dԕ G2s3~l\7aVxJF}>H\]<$]K`Zʸ+'dS~01^_!Fapڐ⃏-Ufl_g_&;҃5Ve% KuȨ^J-Mc)XT=Zs+4ו 0VŧE#h8ilI{[sE `Gy$Ɖ\}+g>6me$U.E͸_$:xX1W%}-!-D(6eY TCmJlƬ=Aȶ6D.*wT Q*#)oҊN?U3ɈN6+}Tȭe+ޑ 姮] /&ͤ ~hѪg;&8GFWʹ6{rwRJ Uچz8+0OSEFGm/1qG|_2u0['L8fXm"n`<!@°514.StFLZQдoaT\ kKmb +"g:|u*0~;ޝ^mߴy8fF *urP[S1?B| W)3N1QE1HF(hWgW_8V0mԾLHKS%J1 CvEHd̟TiAXXERY}cIfJ>\QAˮҶh{VC[ٺzp/"W45Tی=T!vKUJ*Qmʢ{X 4˴?z}f4'Hvfh!`8 hxlgK ߤ70%ўpNPg&3KVVꜹ$6<"hA8AlP+Iqe2DDP$"@^o@qp}aNu:d;eub\z6`M R,V,ma6skh1Gz3 E1x16 $ 1tm v\&BZ)ȏaF8aZc V*q~ҴS{Ah1ò ϻ'{djWjEfW~X6A9[Wrs+0ʔ7MzLK:ڝ7IG_,vpHFySvl[4ΑgDtM JB"Ѽ`=[L-<0YJʽJsIjusrҰp; -96pѲU"FiEek.B)x׺@|u%;M"p5q$&-P <`clSa?~qbXzc*{,"܆]kIW bxT~+pU ){UP2=m,*30zYcSIXU9:۷C-y)c$)ߠ{ `Lthz۶AҘŝi+mq H")TnnAo8Td; THTyC2q=(.B?̱hkո7D!v oKyŃ&b#6*2*-8nt{6|Ǿٻ9ݪP MRa_] ^y͍MEIkc b3B0a"`.9 rםMIR?2+vt TS&EئڲPoWy#.kSKa4,*!ݮa3W?1`P堎R~^2g a8[eŻzIQHtY6%|`$Nx@J yV@]vx&<X 70UciKplbS WF/"&ȠxʗPv+*Kx$4sJPHq*|Fmg_!IU Q$尽Ё&j0)88ڜongj9τ] 4ʕyF|O( ْxMI}>]k-ؓVSشlWgpwp{:x  A]ݧ@`dg5JPq M!_DUHԿڊ~-WIcv~2>e# ^kЗr8Vܡ!+~K/.|i#Ř̍cZ}dN2JQk|^e&k2 t8 D;CTk%Ϥ$P4nJ5_oP~'ڶ5Sc31,'Reۦqu`jJ"XѮH]9(H4,"}&7 `ǯh>F|@dz[ *Qhz?Im C\ZԗCm8*>B<-.%ʅ˽-(ܼڀ$o=o-$H*N\f7E™NP܊eҶ[ #3-I}y{xݬع._ Me5>GX^F+Eʽt0l69ٻ Amh7#mPձI3uP{ÿPwfR!_7w[ LSBFY=:b֢qԱ>DW%ZH?(M52"8{CE$(O{BcY7u2H9%TsU搳U_pFbw 1킧OAq{.ӏWm}5z]k:O!_rB_ζJKdn%tLZ ]EMKQSy痼QZ/-@Tx &=%HΈ#q; ſzי֤=䨏w#(ll,( ,枕K1\h}d*";fIIzҟS^jg:pYm1Bde٩%79N7bgU`|[}3yq9^5 Šur70e)N[ՐtB8/Ask`~,Nˆڎ-8z 5P/}ܢb\nC!a02o >-B|h}d8Μ(Q>M(%*JNxE-nJGñEYLT (lCqX]~V\#lr/.YBt"Ɓ|%Xr <@i&:LUMGZeVEsZcjO iI1q]q o\πEBa!/WrIРX rMt LtU9gvF̟rԑF|2Noy 2 M E] ,lPFe$[Ζ.V\`w\yڥ2g}>˴7wGd1qdiՒt6#%ua/q2+E]WKCo4y>]) V5k1z*>|ua U˼f(AC4?T"29e `(zb!9fp̶!;dt=4d~]h?Ѩj[&/'(N{pMܚ,@]SA~5|qS$O<%דeiw[\ƶZڇ{05B|mkj L^i2Hڂ\"#yh(,$o/}--(A`"@4!qN`tXhZ<ڀ%C.J:p Lz q#dE7{SW=VGdl$@e=ٗTqUxE:b 'IíkXx1@ߜ ф0jB_#yt+\-V7 ^ QU 8>7nn82GMpv ⷣ3l5yRnJ$G?s%]85> el+Kՠ Ojz,d $M5 CB-|lF֍U v $h 6d/ؾۏY7,|Wm<86lO#oS lf)krZs%ygֽgc@E? G_W>'⬘M7Mb cRlQ_ӬioGB``%7JT0d8u^fj )Ht-2awda]J::_-Y2QbA1[<ɠ{{~:8ƛ@@(҇Qr?/e BRn4sC\;xdCZ3bYXjh0#$k!_WvšTH59M5n2/ ECJ !OYJ:Q; /CٹpnZPIƿSeHaέuxMѣ-D̲Uҁ'僧&#Ѡ)KDE& +'1ﮛK-̠I&LI2k3{B,-сp 'L ;86tfB]iö)W{ ڭ!'g9̛c8}w!t~5+nynKNzBX5W٩rSVU;\dp+8ADϬbmI!@}g}̗`> :)yה8{-9@֙0O^ Cl-GqzYEtf~7 VXk"mV㯬iZVz'؍rZX HME>mLJ|5˕(3g8tnϲ8LYܖUFNO[ 2qA_ ,=3 ϖNثLLSHu3y6 a#~sܮ:{*{.`jzRԑ4~?V!!íJDJb3@ĒŶ ZZƗcUw#Rϔ%Huc"% a$ی NF;}j]C}0czW 9{tU]hJ<𽳧m#w:ť-MX/*o,ٞ_.9VܙkW\fGu=]&r"D ߩ2%9qdZc:' Чw)BYh,"u{! 0%~ w&{#0z(brgO)uY,ƶDeAԟm>:{d.}obsA'(zA﵁{$0$O̊D4$;}qi^-.~[tKv U?gQyRU̢| FqY`bGǯ=6ȗzgvW͹NoŅ3^v x$#[~勃OzyXr56y -[L`y+T H?F3Iv G- edXY7C2cox*Bԯ䆷Q0ϢwM ׁsI r [0'BY2[ObWJC<2G&PeB%?Δ3WEIM~yp8&&$YpG> 7 #bW29%=. i'! L(l%S 8y4vq.sɿoW=bi׉˻c* ﻚ"gw :%Ib_?u|7TG;oMu5OUq΀w^O+P3i (x&icWNjvQ£x9&eIg' ̉a!X  B[nkXRlnެ-h_kC$8/#nC 4cV%jU=5+jc?w]r#,۵*k` # <:D!_W&`V\*܅v6=א%`弼ӵ_ի]/\Q1Cxf- kh$ta}|{ Bk,(wKզt4J9\]Xt yW㜉eEk_e>}L/T&د@\C7`> (T+kkv&7k)b90ARZ#dpLәG:b'ȗ~W͢.NZ?.eeL6T]KT>VmAʢ_@<3(CFaJrL`w/GI33*kP9=i'zJ/PR +i~z [{H*H0lK"E<;e6N_JbdЗY;%,ئ l2Xߥ{<:I(]0+:>=دŵ2 Xk~M$ *_9kP~7D4nyUymhaؿg5. ;E6[2-i]>(x3ok;a ,UcZ uGxݲ1!M(ym%犩Y,%?ٿ EK#& ;ׄr"[npXod}R_ЬIV σKslBݭ gl ?Lq$l^2ZiL# u$_X#C/ _+W4=Q3UWI(}|j}3& رsNĈg w!{l5+GIgJ01m|) 1H '3!׊e"ɖ$c} qGGH?NIcd> A.T79@dNfWnN#?MGt_q0QhO'Eƛy8z {R @aQEhשo=83W>5ӍVYi!9 $ɺ|hbZ@kt,WX"?__rAxF_~fhor0SbLsS;%2 -}>JݏÖ8rlk[^Z4rDC:;wn)d$qV)_Mdj/l33'gQoEk(Q=߁sxp HɔHXO% fp(ɗc/gJaTmp7ޥ[.=w_m;Bn u4WZ%U]D@@my7H S;τ&> dsSHsꑿ-5)WWjr36.&Fyq*e|fNIavҔ$7?.%U`APBq<.ȠWVP<h"ߡg[OdvJq;f=ƭ! X?3]7&A ԋtxשEqeKU\8%c~_TE/Tb2d;xՑCJzdA [eo!D)e_1^== _~xrV~)\/;Ni7;cӧ> :^YXrS]3øKw N).½&yܵ6~|1۬avZLq7xEv ^͋_Cb0hĸ*"=jt,gς %)EyWŅ͌lSTBb]#03 "oKnN<'~;̽ ,73-ԡn^(%􄈓ljnOrXM?0\j&\N0~-]Z\=RqK_S~IE<3 HJܐ_0*mpv[+; %0a֤8<ܺlKz:&i jvJb|vX qSth_fKU5A."E\Պ;ŎW6 ߅2̥ueu2$jJpsf@<+A%(^ɾ-jfvkP Q}} 6^=}p!Y0`8tvҮZD-:2Z4zZas=G~hvT@%QSDPif6*~9of7b<}b}Lj m)2E(a?s:1O@F|ɼ*re##߂s̸ &>}|[椀 Iu)Z eЭĦ~͖J}e딓!QÈZ-`IŰ, ^[*ڻQxW}@!)(D ;᝹s2pϓưa BNBQ5O.$&R B_>b ?+q'AcWFe {>uAYqXW68vi)M&mѭ1E8*}% $(12>g40%2B7C8zm`0@mC)##5vO [[;8\Q%92 M8 D)ҥ2Ʌfy1A+wL )c -E6#WF2Dß:hkcI)7_Go@8y#Z1Gͣ *㰊e$央@FH!!ɮr:X}Cel,ϑ=2a}ւ {1gPLσΜQ$[nzfiz%﷼n]ErɹE0;z^kk > 3*\A{!=-ʝX*)t7r*>~=RHI >r)S8.~рj@L#LrIg/Zs(YPnRw巣e>&PbQ3\q`LK:R.;WHw O/?90 @ؐS׈ [G09< WO?U@cfD܄OW蝣ĺJ"VQ04rZڜp2Z~V 1jؖ=vA!9Z2!N"9_R`HKpv m{*zeƌzսaێ 몢887s2iȄ nKAec@,4uXe[0]%U /zq BÖӵx!Kg2? [:X鿩W+ Zϒ>픴\RѰ*m.PNk?'3fhr(E#Bm|{Cz$wvWe$'L?feiSD:i uiaOHL~?gA/}{ U@9; +7D+e'vJ^e\tB<}/ryg1գ|ܜw"'(*͐y_P)v@nmO p80ntDB}A{9G==I҉\ W.SN VEtV7w5WSF[C#fՐ1U֊ $[*$`<`SObo2%ޛ"Ѫݙ#nLof Ugۑ+$%M#z)7SPqu[gҳԞ,>+䰠2sUjPZ6hCN\$ -UvE,fRq!ARZ\څCag>R@z@oUeZ.VЛhoNx_HOUt^ Y?v;i)MJY%~/7*'1(:D(au?F "hsN;FYh-u陊>QlU):=EP]Df<lqLqVof+8k?bmu+1Pcw_v۠yR'LLd'*P`Uh3 dAxzRQ͠\dP06xfծ((w}'!֘MӬ GV[fL@.2'Z^wxy XL/_V)EVDtZr{EUsU~m eR}30 ƺ}saQn:|ŭ6f2 8@>D1jdV.&ҡIO<=9-ȮtkgE=T;(Ꟈd VÍ,`o/z DS(>T6I1=R%2һ~d.7KrZSiT~MB#!~=832>qҷ\}a} 98UZgj+B<8Kɤ }(&TNmVqHIE!e?Ӟu$'L)R"* vt)gšC-@u1.H:]&u˺H, 3CnZ$&~C }ĔlЎ=k M2߻Z YF<&;aV_tVyQ-,ɤq/чVwKZ TƘ@͟ORuIbpqe!F1,O>G /:.w gUVL:Z@N @+,kIRWvݵPv QIlM7@)p?+-0j;ovEHRiO̻YVi4MsœBgk"+dGRͬrF7[{KE6 A(ף+`NaSGCJ~J$=ԣ~^= q:]5^:"0Z8j0A!؝ئEG (lM/{E[J֨D84Ⅼ,52e,.J;sڻ@v}R{}AiC,$d}tp#Y"U͓A_5H{vj|3 ǃvv<#~x-7hXd;Tbm3Ҫ?^'ںTtjm(}`2$UkeO#Cs xS4x2?ӂ V3!缥C ) ~P fT;5c][^)JKǶ|3B/>XSWKX+A-\-egi$%ޏqL5sL6*,s2-Q;ck%W߸aQ~(o}{\z=SULaScY*/˓b{"SA} O >sC+=@  [_8# 'bHHr9 5 􏔒Yt^ i޼šF T[($\(K4AinMhd-:π`-%Qm0OsUykǃnP}{WMEH>w B^σ=Ch+ -8̀wm_D[d/9KdK)3q+HG{뙝@v6%`e9/fzh6]E|hݮ8<@LS@*D / a/Z._J4 {PdJ%&f <3;*4t.&c{*AoԊq t2xy-Vr)ǭp|_21gH$& t_ywR\ y7&mk)f8o'5%L*.qmO5sZǺDJCVwp7Kay0K!]\IH hA힩.<<Ȉ%JL gpN–ʳ%ӤՑck3:{&7~,7fo1oΖDݝ%%-(aO(ǩĹ'9P^]}ro”" z|F!ߐD_ )vNysfhh=Tu"::uR#q*`nD9hj=nml ԫkͱ掗_r\jeeʥW8wRsKpI_P{& HiW|vO0CѲ=_vw@:ҹyxV/nm '`D"Lǂp z"&֚%Rb:?5afIX㞋E7o4u,)kbίɢAn$BoН3TTg㑆OgZ=Pc}*A?;Ȇ%ms3@~?):;xATφĢS<)7ߓӹm)>&fGu>.;3m+D=^$H,Vb.cW= 86p;޳ʃ1nb/Lxa[&W6d  !+zdD4\+S3TEAZA*. *md0f#]Y$ ~IӾnؐ(HA#RzzѸ2Ɇ$$ޢQW8♂9咧lօjd+Հaշ<<込)1Q6Jd*[gxmqj| ]i8?.a-/ai0dY78j?xg09A/7q(6YiLK0)_*A/V G`'[|E5W'/Ϟrp+qBt!˓amT~92MLVcGiՐAbs9}'հ8hѥhaӫc[7k)(uX3+=4 Fbr=6r%zx^L猖]`Z%D/{\2ޕr5H5?t PyX(GuRؽY%illKcj<9 :}//,(*dF&;kKAS:~?Y$37=>߿ĸ@!6&_>G;j XqE^zaa%p !Αb+߷JX:Ro&;ꅻR[ʓPMIak@ F=d>֨))zGi+a!5 [/yf"f6|[#5phL'~,`\;IT-:m.s3 ?v &ѧHaz4dus7Ӟ]ٿD{s>}&HBC?fSp;gIo)bÛ%2@k-\t})z+N:^irSɐS}ys|-^쇚*IMuw4AZq#P]A3^)`9]d鷈Y[Tgw/E[(Gom91yPK?Y9%eR ŏQ(ή.u=Q 4x$E$Txt1jRǒz+9Ȩk ycZ87r~ivTi֌%e(gr  }, /3mտ吆$evWRNK1#B9YOiH U%|$R${]E8^Jc_0#>yNH  zMyNXXvVXQBh Y4 }h{rws+9Z$|jf{7Ē 0PC>0ɴm\/p~J02'sPZ=ș-bVX#(qPb8_ ,X@{ c֔PpHaL[t Fof&ȉ!QE&(0@̯:8т}NHSX1U50%`= A9B>1v#˽w! O3|=_?""2ǐD !9<J#q̓BEȈ+D _7]^'<@.+fH"G0iϠ~Ħ7?$mj:6 #O}T6 %v= ~NUdna/ I za)hn!*F1ݎqb,@>5B,ÐE dʯ%/MYlJ4ߗ-& #8dl|, >0 f,om>`>qWG۸7Rk:;6{'D ybA5o&t}c&/d}f~C7cJ%>OwԀ0}(VJ]Ld,W:4R,!wUywB>P3t" c"vc]^x25quB19f#/}+]\gxgb>tJjNatGt?J2} ߔB }g9qd@WGӞq>\guJ|BĢ 88_FIg*(= $ |s^0`L\q92_.9rFEEbͼV2‹۪A$Eiy~a0w #]]¼'/|}=`LƋR*yGt1(Ù<y49  G?jS8o~%L-pb-#1zgyiCZ#YQ8p.T|1k7ҫ]ӏΘ~Sb^W,S0(Z4a3Rn7I?Y_hR9?V`)w![`8Oi؜5J0ܬRfs>Nc VWe,ҏ TKlsѠ R7-B(גA9EZ1]O:w!8Tn&fH&PyLƫml^:V*+{D!uD-qg0ϻ3ES%ưA(yh;Ŧ$sdս]@M z|@f7C|M3%(w(.4:y=R=1 #Uźg-/D)ޛ9D@X_yunND=Q^ myaשHG=ecT眉Fj L?u%4,t8#U }.U4 zv5*4&6OAOCv0Jo!=T1}Fܼ{9E[1V el[80SuE\;DTֻhD&3>)kU!7j _{_ vyjGІűkbD}O0jzѥz)iXx:9)7)oF&dg`h`KxeF;swzQ:4y_NtH-q0 \g\G/Hʂ: }+G*\@}MF?CҞi`e3d;*=3@*K,,-D1hzG:ehl&7CBUXx:5Rz+Y+eS6ӥN"^o'&!凪ٴ>`TR?[Xǜ/@vayQ_9~i]7W Ͽk,!ňh*p&yLc ޝ >N!c!$UAM)ٞ` j{% ?Yұ^61kdX,nQd?3/gAF>t2f+%8!I؞.Uܣ']1⟿'тXUSk{t 5hqa\ 340Vx}u3_%Ba&4WLA%6H$[1SՒ <VWg4/ ̺ӄbD$Ǫ2 $Y~sWӶ0K#!zO=Rh^qXi*d]X:0kb!ɳi?.+;JQsnA[n%TO_鬴b V?4':tC˨Ggu%b^@1@n]Gq%Y!_k*`v]ÞU4:dڋFNBQ;y0d~-Dd8t6B|l:TN d0Z5:L*`aas.$y'CvhF6536_b[̦ܳmeٖ1F'Px2P:_/"W&A#GbXw@z9v~ڧOYFM@3 B@Bk(H=^ױ %{w2e 0~JOu g]4_6aHΦNveAQ|^Ff˚.|3daohKb? j\F!y#S,9\o*[A2{e)r~>j:,;)# qԌŭ;2OQd#I{]Cd2BrŞ8=,3Z&r t֗SŐ*YƂq2݊diu5*o]t3Us2 7\Zވ-|9 v  e:&䒮|E$\zn*qttA<+4?/+4b-Au/? xQ4=*J! ,ĆH4pcH.㋯ P=@QlrϰfaYq*vHt"p~_ %sBzq< W^)h +T0śdƗ'jĦz<yLo㟞:ؾ#&lhZKf"\]u?ͽXvKHO~IjEm΃M;: cll@4Bs mxˆx>LPR!xDu;}~o|sԼ!?nx2`NO:>)fm8sˋ3RxA+Bd aa YZ