Re: [ACPI] New ACPI source release (20010201)

[<ti...@ja...>]

Let's test if the patch actually compiles before releasing it, huh?
I realize that Intel employees are not allowed to keep any "Linux"
machines on the internal network, but cmon, this is ridiculous.

* "_LINUX" gets redefined 203952358 times in acpi headers
* final link fails with
drivers/acpi/acpi.o: In function `acpi_battery_read_info`:
drivers/acpi/acpi.o(.text+0x284ee): undefined reference to `__ucmpdi2`
drivers/acpi/acpi.o(.text+0x28507): undefined reference to `__ucmpdi2`

I am compiling with gcc 2.95.3, not with some redhat compiler.
I had this problem with another ACPI release earlier. Obviously, nobody
bothered to tell me what caused it or how to avoid it. See archives to
find out how I got rid of the compile error (obviously not the way to do
it).

And yeah, after it compiles, on reboot I get a oops very early in the boot
process. I guess ACPI dies. Just to make sure I untarred a clean 2.5.1,
patched to 2.5.2, 2.5.3, applied acpi patch, run make menuconfig, make
dep, make bzImage modules modules_install.

So, no 2.5.3 with ACPI for me.

tim

-- 
・‥…━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━…‥・
     timecop at japan.co.jp | ＯＡ通信サービス株式会社 | NTT DoCoMo
      Please don't CC me on anything sent to mailing lists or send
         me email directly unless it's a privacy issue, thanks.
・‥…━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━…‥・