home.social

Search

176 results for “denzilferreira”

  1. I've modded my Thinkpad T495 by installing an NVME on the WWAN M.2 2248 slot. I used an M.2 2230 nvme WD Black SN770M + an 2230 to 2248 adapter. It is rare and difficult to get 2248 nvme's, but due to Steam Deck, it is now easy to get 2230 nvme drives. I have now 2TB, 1 TB on the 2280 slot, 1 TB on the 2248 slot.

    For Linux to detect it, I had to boot with these kernel params:
    "pcie_aspm=off nvme_core.default_ps_max_latency_us=0"

    I also now have upgraded the factory nvme firmware following this guide: community.wd.com/t/firmware-up

    Here is the link to download the latest firmware: wddashboarddownloads.wdc.com/w

    It's not the fastest drive, but I'll take it over having an external USB powered solution.

    #thinkpad #t495 #wwan #modding

  2. I've modded my Thinkpad T495 by installing an NVME on the WWAN M.2 2248 slot. I used an M.2 2230 nvme WD Black SN770M + an 2230 to 2248 adapter. It is rare and difficult to get 2248 nvme's, but due to Steam Deck, it is now easy to get 2230 nvme drives. I have now 2TB, 1 TB on the 2280 slot, 1 TB on the 2248 slot.

    For Linux to detect it, I had to boot with these kernel params:
    "pcie_aspm=off nvme_core.default_ps_max_latency_us=0"

    I also now have upgraded the factory nvme firmware following this guide: community.wd.com/t/firmware-up

    Here is the link to download the latest firmware: wddashboarddownloads.wdc.com/w

    It's not the fastest drive, but I'll take it over having an external USB powered solution.

    #thinkpad #t495 #wwan #modding

  3. I've modded my Thinkpad T495 by installing an NVME on the WWAN M.2 2248 slot. I used an M.2 2230 nvme WD Black SN770M + an 2230 to 2248 adapter. It is rare and difficult to get 2248 nvme's, but due to Steam Deck, it is now easy to get 2230 nvme drives. I have now 2TB, 1 TB on the 2280 slot, 1 TB on the 2248 slot.

    For Linux to detect it, I had to boot with these kernel params:
    "pcie_aspm=off nvme_core.default_ps_max_latency_us=0"

    I also now have upgraded the factory nvme firmware following this guide: community.wd.com/t/firmware-up

    Here is the link to download the latest firmware: wddashboarddownloads.wdc.com/w

    It's not the fastest drive, but I'll take it over having an external USB powered solution.

  4. Been three weeks coffee free. First week was hard, but now I can feel the difference. I don't feel tired anymore after lunch, or afternoon. In the evening, I feel it's time to go sleep (around 11pm). I wake up without alarm at 6h30 and not sluggish. I found my circadian rhythm!!!

    #noCoffee #circadian #alarmless

  5. My son won the player of the game award today at ice hockey. On the way to the lockers, he looked at the goalie and gave him the award: “Without you, we would have lost. You were the best today!” - so proud of him!
    #kkp #icehockey #suomi

  6. Part of my job is to step back and slow down. It's a skill that I learned during my frustrating years as a researcher, when we are pioneering something and there is no easily accessible answers to our questions. Now in industry, I find a few of my coworkers struggling with a problem for days sometimes. If this is you, reach out to our coworkers. More often than not, someone has seen what you are seeing and can definitely help you.

    #imposter #DevTips #SeniorDeveloper

  7. Then your Docker compose container should have:

    image-name:
    build:
    context: .
    devices:
    - /dev/dri
    - /dev/kfd
    group_add:
    - video
    shm_size: 4G
    environment:
    - PYTORCH_HIP_ALLOC_CONF=garbage_collection_threshold:0.9,max_split_size_mb:512
    - PYTORCH_CUDA_ALLOC_CONF=garbage_collection_threshold:0.9,max_split_size_mb:512
    - HSA_OVERRIDE_GFX_VERSION=11.0.0

    #rocm #gfx1103 #780M #radeon #amd #docker

  8. Then your Docker compose container should have:

    image-name:
    build:
    context: .
    devices:
    - /dev/dri
    - /dev/kfd
    group_add:
    - video
    shm_size: 4G
    environment:
    - PYTORCH_HIP_ALLOC_CONF=garbage_collection_threshold:0.9,max_split_size_mb:512
    - PYTORCH_CUDA_ALLOC_CONF=garbage_collection_threshold:0.9,max_split_size_mb:512
    - HSA_OVERRIDE_GFX_VERSION=11.0.0

    #rocm #gfx1103 #780M #radeon #amd #docker

  9. Then your Docker compose container should have:

    image-name:
    build:
    context: .
    devices:
    - /dev/dri
    - /dev/kfd
    group_add:
    - video
    shm_size: 4G
    environment:
    - PYTORCH_HIP_ALLOC_CONF=garbage_collection_threshold:0.9,max_split_size_mb:512
    - PYTORCH_CUDA_ALLOC_CONF=garbage_collection_threshold:0.9,max_split_size_mb:512
    - HSA_OVERRIDE_GFX_VERSION=11.0.0

    #rocm #gfx1103 #780M #radeon #amd #docker

  10. Then your Docker compose container should have:

    image-name:
    build:
    context: .
    devices:
    - /dev/dri
    - /dev/kfd
    group_add:
    - video
    shm_size: 4G
    environment:
    - PYTORCH_HIP_ALLOC_CONF=garbage_collection_threshold:0.9,max_split_size_mb:512
    - PYTORCH_CUDA_ALLOC_CONF=garbage_collection_threshold:0.9,max_split_size_mb:512
    - HSA_OVERRIDE_GFX_VERSION=11.0.0

  11. Look who is here? Pulled a few tidbits from Bigme Hibreak Color phone to decompile to reverse engineer how the OS and apps interact with the eink display.

    #eink #bigme #decompile

  12. What is Adobe doing with Adobe Reader when its download size is close to 900MB?!

    #adobe #CodeBloat

  13. The smell of gingerbread cookies out of the oven is only matched to bread loaf of sourdough. Love it! #baking #gingerbread #HolidaySeason our own recipe of dough, very light on your stomach.

  14. My Pebble order is now confirmed and shipping soon 🤩 excited to go e-paper ~~e-ink~~ on my wrist! #pebble #epaper

  15. My Pebble order is now confirmed and shipping soon 🤩 excited to go e-paper ~~e-ink~~ on my wrist! #pebble #epaper

  16. My Pebble order is now confirmed and shipping soon 🤩 excited to go e-paper ~~e-ink~~ on my wrist! #pebble #epaper

  17. My Pebble order is now confirmed and shipping soon 🤩 excited to go e-paper ~~e-ink~~ on my wrist!

  18. The beautiful HP Laserjet 1010, easy to maintain and fix yourself. Fuse easily accessible 💪 and markers on the printer to let you know where screws go into. It's the little details that makes this printer a solid performer for more than 20 years!

    #hp #laserjet #oldschool #printers

  19. Kuuma in Oulu today 🤩 was a fantastic gig! Something about how Finnish is a great language for music 🎶 #kuuma #oulu

  20. My dad taught me, I'm teaching my eldest to know how to take care of their bikes 😉

    #biker #family #nextgen

  21. We're expanding our R&D team in Finland and we're looking for Android System Engineers. If you are interested in learning how to compile, change, and deploy your own Android OS, develop and ideate on system applications our customers will enjoy using worldwide and build future technologies from scratch, this is for you.

    If you or, someone you know is interested on joining us at HMD, tag them, reach out to me.

    PS: This position is Europe-based and you can work remotely from Europe. Most of the team is in Finland.

    #job #hiring #hmd #engineer #developer #mobile #android #aosp #hired #fedihire

  22. Let your inner child free for maximum enjoyment 😉 #angry #birds #rovio

  23. Want to create your own AI agents, wrapped on a container to deploy anywhere? Allowing you to test locally before publishing it? #daggerIO #AI #containers #agents

    docs.dagger.io/ai-agents/

  24. After some trial and error, I now have AMD GPU accelerated LLMs running locally on my Ollama instance. I have an external 5700XT (gfx1010), which is not officially supported by ROCm. However, RDNA2 gfx1030 is supported.

    So if you want to have a 5700XT working you need to:
    - Install: sudo dnf rocblas
    - Go to: /usr/lib64/rocblas/library
    - Execute: sudo ln -s TensileLibrary_lazy_gfx1030.dat TensileLibrary_lazy_gfx1010.dat
    - Edit systemd Ollama.service to let it know you have a 10.1.0 (that's what the gfx1010 stands for) via an environment variable by editing the file sudo nano /etc/systemd/system/ollama.service

    It should look like this:
    [Service]
    Environment="HSA_OVERRIDE_GFX_VERSION=10.1.0"

    What is cool is that the iGPU is also loaded up and it's CU cores and VRAM are available too!

    #ollama #5700XT #amdgpu #llm #opensource

  25. #housemarque #saros grabbed a physical copy. After #Returnal I'm really looking for this masterpiece! @bonzrat did you write the storyline? 👀

  26. #housemarque #saros grabbed a physical copy. After #Returnal I'm really looking for this masterpiece! @bonzrat did you write the storyline? 👀

  27. #housemarque #saros grabbed a physical copy. After #Returnal I'm really looking for this masterpiece! @bonzrat did you write the storyline? 👀

  28. grabbed a physical copy. After I'm really looking for this masterpiece! @bonzrat did you write the storyline? 👀