There have been two PAN fixes since the patch that introduced PAN: * 6b6ac525f7bcb4259ef2065919bc9985c48ef35d * 0e4c3437fdfb20e0f0d590e0d0abe32fc8393b3c
Can you try again with those? ________________________________ From: gem5-dev <[email protected]> on behalf of Gabe Black <[email protected]> Sent: Thursday, September 26, 2019 6:47 AM To: gem5 Developer List <[email protected]> Subject: [gem5-dev] Trouble booting Android N on ARM with PAN enabled Hi folks. Recently we (google) stopped being able to boot Android N on gem5 after rebasing our internal branch. I've bisected down to the offending patch which is the one that adds PAN support to ARM. 361bc8b47260a55902764054d3ac25694ac93f8a I can get android to boot again by either reverting this change, or disabling the two config bits which are changed in that CL and default to on. Does anybody know what's going on here? Is the implementation of PAN incorrect? Is Android doing something wrong? One thing I noticed was that that CL stopped clearing the 22 and 23 bits of the SPSR when taking exceptions. I sort of see why since one of those bits is now the "pan" bit, but the other is (as far as I know) still reserved. Should it still be cleared? I tried adding back in a bitfield for the appropriate bit and clearing it like before, but this didn't fix booting. Gabe _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
