On Sun, Jul 22, 2018 at 11:19:00AM -0600, Theo de Raadt wrote: > This is unreasonable. > > 1. Can I add High-performance to every 2nd line in this file? After all, > I bet everyone would love to claim that for their hardware.
I'm sure. That's the only way the generation one and two hardware are described. How about not making any differentiation between first and second generation in the string at least. > 2. These product strings get embedded *into the kernel*, bloating it. > > 3. As a result, we use short boring names, or highly descriptive > product-serial names which won't bloat the kernel. > > Your diff is simply wasteful. Fair enough. How about this or is this still too long? Bryan Index: sys/dev/pci/pcidevs =================================================================== RCS file: /cvs/src/sys/dev/pci/pcidevs,v retrieving revision 1.1854 diff -u -p -r1.1854 pcidevs --- sys/dev/pci/pcidevs 8 Jul 2018 17:34:07 -0000 1.1854 +++ sys/dev/pci/pcidevs 20 Jul 2018 19:39:47 -0000 @@ -279,6 +279,7 @@ vendor BLUESTEEL 0x15ab Bluesteel vendor VMWARE 0x15ad VMware vendor ZOLTRIX 0x15b0 Zoltrix vendor MELLANOX 0x15b3 Mellanox +vendor SANDISK 0x15b7 SanDisk vendor AGILENT 0x15bc Agilent vendor QUICKNET 0x15e2 Quicknet Technologies vendor NDC 0x15e8 National Datacomm @@ -6975,6 +6976,10 @@ product SANGOMA A10X 0x0300 A10x product DIGI NEO4 0x00b0 Neo-4 product DIGI NEO8 0x00b1 Neo-8 product DIGI NEO8_PCIE 0x00f0 Neo-8 + +/* SanDisk (Western Digital) */ +product SANDISK WDSXXXG1X0C 0x5001 WD Black NVMe SSD +product SANDISK WDSXXXG2X0C 0x5002 WD Black NVMe SSD /* Schneider & Koch (SysKonnect) */ product SCHNEIDERKOCH FDDI 0x4000 FDDI