XPost: linux.debian.kernel   
   From: ukleinek@debian.org   
      
   Control: forwarded -1 https://lore.kernel.org/all/aZMt6sSlpKK_JAdT@monoceros   
      
   Hello Paul,   
      
   On Tue, Feb 10, 2026 at 12:19:45PM +0000, Paul Kerry wrote:   
   > Subject: linux-image-6.12.69+deb13-amd64: hp_bioscfg mm/page_alloc.c:4802   
   warning   
   > Package: src:linux   
   > Version: 6.12.69-1   
   > Severity: normal   
   > X-Debbugs-Cc: p.kerry@sheffield.ac.uk   
   > Dear Maintainer,   
   > *** Reporter, please consider answering these questions, where appropriate   
   ***   
   >    * What led up to the situation?   
   > Booting a system into the recent debian 6.12.69+deb13-amd64 kernel upgrade   
   from 6.12.63+deb13-amd64   
      
   Probably the relevant difference between these two kernel versions is   
   commit b2e79963525205707c87391a01435cfdc9fc2d1e   
   (https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.gi   
   /commit/?h=linux-6.12.y&id=b2e79963525205707c87391a01435cfdc9fc2d1e).   
      
   Can you confirm that when booting the older kernel, running   
      
   	modprobe hp_bioscfg   
      
   will break in the same way as the new kernel? If so, the driver was   
   broken already before and just unused. You can achieve this behaviour   
   with 6.12.69, too, by blacklisting the hp_bioscfg module. Either by   
   adding:   
      
   	module_blacklist=hp_bioscfg   
      
   on the kernel command line, or   
      
   	echo blacklist hp_bioscfg >> /etc/modprobe.d/broken-hp-bioscfg.conf   
   	update-initramfs -k all -u   
      
   >    * What exactly did you do (or not do) that was effective (or   
   >      ineffective)?   
   > n/a   
   >    * What was the outcome of this action?   
   > dmesg under 6.12.69+deb13-amd64 is now displaying the following...   
   > [    3.346704] ------------[ cut here ]------------   
   > [    3.347194] WARNING: CPU: 11 PID: 606 at mm/page_alloc.c:4802   
   __alloc_pages_noprof+0x2a7/0x310   
      
   This is   
      
           if (WARN_ON_ONCE_GFP(order > MAX_PAGE_ORDER, gfp))   
                   return NULL;   
      
   which is probably triggered by   
      
           bioscfg_drv.enumeration_data = kcalloc(bioscfg_drv.enume   
   ation_instances_count,   
                                                  sizeof(*bioscfg_d   
   v.enumeration_data), GFP_KERNEL);   
      
   I manually determined sizeof(*bioscfg_drv.enumeration_data) to be around   
   20K. I have no idea about what the driver does here, but it seems to   
   request more than 2**10 pages continuous pages here which suggests that   
   bioscfg_drv.enumeration_instances_count, is bigger than 200.   
      
   @Jorge: Is that expected?   
      
   I dropped much context (mostly automatically collected information about   
   the machine where this triggered). If you're intested in that, check out   
   https://bugs.debian.org/1127612.   
      
   Best regards   
   Uwe   
      
   -----BEGIN PGP SIGNATURE-----   
      
   iQEzBAABCgAdFiEEP4GsaTp6HlmJrf7Tj4D7WH0S/k4FAmmTOOIACgkQj4D7WH0S   
   /k655gf7BYwadASf0/z9WnkbUgSemPpovHXczhrSVMORilULBVZntwMijweu1QQJ   
   B4JDDRu777kdrkfLcW13LKDuzBa64bFWGWSv0qk9xQuF28CVLKVj+z2VjSzGR1fV   
   l9rVV51pw7fvRtHdg5T66dqTsORILC1hpHxW/CXIFGThhTNfV+bDKITDAFYAvjpC   
   1lDuUtD3yhpfJZKKGqBEKtDNqDBzCAQ1l44oEEn70ARs5EEhAx+Ob7XBx6bguWU/   
   x3Vz7HeHvsBlk6ksDEqCniwwm59/v8fGQx6/UVSd2/Qge4nZFWZk+Wc0LI0SKKgK   
   tw2l414dKBpzZK9IBADzugCRd89t0A==   
   =3xX1   
   -----END PGP SIGNATURE-----   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)