{"id":3400,"date":"2025-12-29T00:20:31","date_gmt":"2025-12-29T08:20:31","guid":{"rendered":"https:\/\/rglinuxtech.com\/?p=3400"},"modified":"2025-12-29T00:20:31","modified_gmt":"2025-12-29T08:20:31","slug":"nvidia-nasty-hack-time-again-for-gtx-gpus-with-kernel-6-19","status":"publish","type":"post","link":"https:\/\/rglinuxtech.com\/?p=3400","title":{"rendered":"NVIDIA – Nasty Hack Time Again – for GTX GPUs with Kernel 6.19….."},"content":{"rendered":"

This is a summary of the changes I made to the NVIDIA proprietary ‘kernel’ code to get driver 580.119.02<\/strong> working with Kernel 6.19-rc3<\/strong>..
\nThis is a hack, and is not made by a professional, so use at your own risk, and do not use in a business-critical environment!<\/strong><\/em><\/p>\n

As mentioned in the previous article, I have tested this with Fedora 43 KDE\/Plasma running VMware Workstation 25.0.0 with a Win10 Guest, and all seems to work as it should… These changes do not include complete kernel version tests, so the resulting driver is only for the 6.19.x kernel, but does support the older GTX GPUs… Some of the initial changes are similar to those for the ‘kernel-open’ code..<\/p>\n

I found a suitable patch for the GPL issue – dating back to kernel 6.15, at:
\nhttps:\/\/github.com\/CachyOS\/CachyOS-PKGBUILDS\/commit\/b85953540b1b6ee604060870d45a0da8ab3e0487<\/a><\/p>\n

and incorporated that with the following code changes..<\/p>\n

~\/kernel\/common\/inc\/nv-mm.h<\/strong><\/span>
\naround line 204:
\nThis already includes the ‘correct’ code, but the obscure test around it now fails, exposing the ‘incorrect’ code, so remove it, leaving the correct code only..<\/em>
\nRemove
\n#if !NV_CAN_CALL_VMA_START_WRITE<\/span><\/strong>
\n\u00a0nv_vma_start_write(vma);<\/span><\/strong>
\n\u00a0ACCESS_PRIVATE(vma, __vm_flags) |= flags;<\/span><\/strong>
\n#elif defined(NV_VM_AREA_STRUCT_HAS_CONST_VM_FLAGS)<\/span><\/strong>
\nand
\n#else<\/strong><\/span>
\n\u00a0vma->vm_flags |= flags;<\/strong><\/span>
\n#endif<\/strong><\/span>
\nleaving just
\nvm_flags_set(vma, flags);<\/strong><\/span><\/p>\n

– and do the same to the next (similar) code, leaving just
\nvm_flags_clear(vma, flags);<\/strong><\/span><\/p>\n

~\/kernel\/nvidia\/nv-dma.c<\/strong><\/span>
\naround line 718:
\nchange:
\nreturn (ops->map_resource != NULL);<\/strong><\/span>
\nto:
\nreturn NV_TRUE;<\/strong><\/span><\/p>\n

~\/kernel\/nvidia\/nv-mmap.c
\n<\/strong><\/span>around line 20,add:
\n#include <linux\/version.h>
\n<\/strong><\/span><\/p>\n

and after the #include,s, add:
\n#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 19, 0)<\/strong><\/span>
\n#define nv_vm_flags_set(v, f) vm_flags_reset((v), (v)->vm_flags | (f))<\/strong><\/span>
\n#define nv_vm_flags_clear(v, f) vm_flags_reset((v), (v)->vm_flags & ~(f))<\/strong><\/span>
\n#endif<\/strong><\/span><\/p>\n

~\/kernel\/nvidia-drm\/nvidia-drm-gem.c<\/strong><\/span><\/p>\n

around line 45, add:
\n#include <linux\/version.h><\/strong><\/span><\/p>\n

#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 19, 0)<\/span><\/strong>
\n#define nv_vm_flags_set(v, f) vm_flags_reset((v), (v)->vm_flags | (f))<\/span><\/strong>
\n#define nv_vm_flags_clear(v, f) vm_flags_reset((v), (v)->vm_flags & ~(f))<\/span><\/strong>
\n#endif<\/span><\/strong><\/p>\n

~\/kernel\/nvidia-drm\/nvidia-drm-gem-user-memory.c<\/strong><\/span><\/p>\n

around line 42, after the #endif, add:
\n#include <linux\/version.h><\/strong><\/span><\/p>\n

#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 19, 0)<\/strong><\/span>
\n#define nv_vm_flags_set(v, f) vm_flags_reset((v), (v)->vm_flags | (f))<\/strong><\/span>
\n#define nv_vm_flags_clear(v, f) vm_flags_reset((v), (v)->vm_flags & ~(f))<\/strong><\/span>
\n#endif<\/strong><\/span><\/p>\n

~\/kernel\/nvidia-uvm\/uvm.c<\/strong><\/span><\/p>\n

around line 22, add:
\n#include <linux\/version.h><\/strong><\/span><\/p>\n

around line 43, add:
\n#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 19, 0)<\/strong><\/span>
\n#define nv_vm_flags_set(v, f) vm_flags_reset((v), (v)->vm_flags | (f))<\/strong><\/span>
\n#define nv_vm_flags_clear(v, f) vm_flags_reset((v), (v)->vm_flags & ~(f))<\/strong><\/span>
\n#endif<\/strong><\/span><\/p>\n

The result:
\n………<\/strong><\/span>
\nCC [M] nvidia.mod.o<\/strong><\/span>
\nCC [M] .module-common.o<\/strong><\/span>
\nLD [M] nvidia.ko<\/strong><\/span>
\nCC [M] nvidia-uvm.mod.o<\/strong><\/span>
\nLD [M] nvidia-uvm.ko<\/strong><\/span>
\nCC [M] nvidia-modeset.mod.o<\/strong><\/span>
\nLD [M] nvidia-modeset.ko<\/strong><\/span>
\nCC [M] nvidia-drm.mod.o<\/strong><\/span>
\nLD [M] nvidia-drm.ko<\/strong><\/span>
\nCC [M] nvidia-peermem.mod.o<\/strong><\/span>
\nLD [M] nvidia-peermem.ko<\/strong><\/span>
\nmake[2]: Leaving directory ‘\/home\/rgadsdon\/NVIDIA-Linux-x86_64-580.119.02-custom\/kernel’<\/strong><\/span>
\nmake[1]: Leaving directory ‘\/usr\/src\/linux-6.19-rc3’<\/strong><\/span><\/p>\n

and:<\/p>\n

# uname -a<\/span><\/strong>
\nLinux rgxeonl 6.19.0-rc3 #1 SMP PREEMPT_DYNAMIC Sun Dec 28 16:41:08 PST 2025 x86_64 GNU\/Linux<\/span><\/strong>
\n
\n# nvidia-smi<\/strong><\/span>
\nSun Dec 28 17:28:34 2025<\/strong><\/span>
\n+-----------------------------------------------------------------------------------------+<\/strong><\/span>
\n| NVIDIA-SMI 580.119.02 Driver Version: 580.119.02 CUDA Version: 13.0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0|<\/strong><\/span>
\n+-----------------------------------------+------------------------+----------------------+<\/strong><\/span>
\n| GPU Name Persistence-M\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 | Bus-Id Disp.A\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 | Volatile Uncorr. ECC |<\/strong><\/span>
\n| Fan Temp Perf Pwr:Usage\/Cap\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0| Memory-Usage\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0| GPU-Util Compute M.\u00a0 |<\/strong><\/span>
\n|\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0|\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 | MIG M.\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0|<\/strong><\/span>
\n|=========================================+========================+======================|<\/strong><\/span>
\n| 0 NVIDIA GeForce GTX 1050 Ti Off\u00a0 \u00a0 \u00a0 \u00a0 | 00000000:03:00.0 On\u00a0 \u00a0 | N\/A\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 |<\/strong><\/span>
\n| 46% 32C P0 N\/A \/ 75W\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 | 15MiB \/ 4096MiB\u00a0 \u00a0 \u00a0 \u00a0 | 0% Default\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0|<\/strong><\/span>
\n|\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0|\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 | N\/A\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 |<\/strong><\/span>
\n+-----------------------------------------+------------------------+----------------------+<\/strong><\/span>
\n<\/code>
\nAs before, this is an unsophisticated hack, so hopefully someone who really knows what they are doing can produce a ‘proper’ patch!<\/p>\n

Robert Gadsdon. December 28th 2025.<\/strong><\/em><\/p>\n","protected":false},"excerpt":{"rendered":"

This is a summary of the changes I made to the NVIDIA proprietary ‘kernel’ code to get driver 580.119.02 working with Kernel 6.19-rc3.. This is a hack, and is not made by a professional, so use at your own risk, and do not use in a business-critical environment! As mentioned in the previous article, I have tested …<\/span> Continue reading →<\/span><\/a><\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"footnotes":""},"categories":[5,11,14,15,18,19],"tags":[2791,2800,182,2801,277],"class_list":["post-3400","post","type-post","status-publish","format-standard","hentry","category-compilation","category-hacks","category-kernel","category-linux-2","category-nvidia","category-opinion","tag-580-119-02","tag-gtx-gpus","tag-hack","tag-kernel-6-19-support","tag-nvidia-2"],"_links":{"self":[{"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=\/wp\/v2\/posts\/3400","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=3400"}],"version-history":[{"count":3,"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=\/wp\/v2\/posts\/3400\/revisions"}],"predecessor-version":[{"id":3403,"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=\/wp\/v2\/posts\/3400\/revisions\/3403"}],"wp:attachment":[{"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3400"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3400"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/rglinuxtech.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3400"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}