On Fri, Aug 23, 2019 at 04:25:16AM +0300, Jarkko Sakkinen wrote: > On Thu, Aug 22, 2019 at 08:00:15PM +0300, Jarkko Sakkinen wrote: > > On Thu, 2019-08-22 at 07:50 -0700, Sean Christopherson wrote: > > > On Thu, Aug 22, 2019 at 05:37:18PM +0300, Jarkko Sakkinen wrote: > > > > On Mon, 2019-08-12 at 18:12 -0700, Sean Christopherson wrote: > > > > > Invoke EADD with the userspace source address instead of first copying > > > > > the data to a kernel page to avoid the overhead of alloc_page() and > > > > > copy_from_user(). > > > > > > > > > > Suggested-by: Andy Lutomirski <luto@xxxxxxxxxx> > > > > > Signed-off-by: Sean Christopherson <sean.j.christopherson@xxxxxxxxx> > > > > > > > > NAK because takes away TCS validation and the commit message > > > > does not give any reasoning for doing that. > > > > > > Doh, I have a thorough explanation, but apparently it never made it from > > > my head to the changelog. I'll send v2 as a standalone patch. > > > > Yeah, w/o explanation I won't just take away functionality :-) > > I came to realize that also from security perspective it might be > helpful to EADD, not from a copy of the source, but from the > actual source. > > So yes, I'm for not supporting copy approach at all. I think this > viewpoint is important to note in addition to the performance > perspective. Side topic, I'm getting ELDU MAC failures on master. Are there any known regressions, or should I start debugging? [ 18.005880] ------------[ cut here ]------------ [ 18.005884] sgx: ELDU returned 9 (0x9) [ 18.005906] WARNING: CPU: 4 PID: 1142 at /home/sean/go/src/kernel.org/linux/arch/x86/kernel/cpu/sgx/encl.c:50 sgx_encl_eldu+0x363/0x390 [ 18.005907] Modules linked in: [ 18.005913] CPU: 4 PID: 1142 Comm: lsdt Tainted: G W 5.3.0-rc3+ #725 [ 18.005914] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015 [ 18.005916] RIP: 0010:sgx_encl_eldu+0x363/0x390 [ 18.005917] Code: 75 11 f7 c3 00 00 00 40 ba f2 ff ff ff 0f 85 d4 fe ff ff 89 d9 89 da 48 c7 c6 4e af db 81 48 c7 c7 53 af db 81 e8 8d 1f 03 00 <0f> 0b ba f2 ff ff ff e9 b1 fe ff ff 31 c0 48 c7 c7 40 10 53 82 e9 [ 18.005917] RSP: 0000:ffffc90000703d00 EFLAGS: 00010286 [ 18.005918] RAX: 0000000000000000 RBX: 0000000000000009 RCX: 0000000000000006 [ 18.005918] RDX: 0000000000000007 RSI: 0000000000000082 RDI: ffff88846f816560 [ 18.005919] RBP: ffffc90000703d90 R08: 00000000000002da R09: 0000000000000004 [ 18.005919] R10: 0000000000000000 R11: 0000000000000001 R12: ffff88846d08a100 [ 18.005920] R13: ffffea00118daa40 R14: ffff888467d3e740 R15: ffffea001189c680 [ 18.005920] FS: 00007fa90afff700(0000) GS:ffff88846f800000(0000) knlGS:0000000000000000 [ 18.005922] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 18.005923] CR2: 00007fa8e3001000 CR3: 0000000466389002 CR4: 0000000000360ee0 [ 18.005923] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 18.005923] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 18.005924] Call Trace: [ 18.005934] ? sgx_encl_load_page.part.15+0x56/0x90 [ 18.005935] sgx_encl_load_page.part.15+0x56/0x90 [ 18.005936] sgx_vma_fault+0x84/0xf0 [ 18.005941] __do_fault+0x4f/0x87 [ 18.005943] __handle_mm_fault+0xa65/0x1020 [ 18.005945] handle_mm_fault+0xb0/0x1f0 [ 18.005946] __do_page_fault+0x231/0x4d0 [ 18.005951] async_page_fault+0x2f/0x40 [ 18.005955] RIP: 0033:0x7ffe7bd62a39 [ 18.005955] Code: 74 05 c1 e8 0c 89 06 31 c0 5d c3 90 90 90 90 90 90 55 48 89 e5 83 f8 02 72 67 83 f8 03 77 62 48 8b 5d 10 48 8d 0d 00 00 00 00 <0f> 01 d7 31 db 48 8b 4d 18 e3 10 89 01 73 0c 66 89 79 04 66 89 71 [ 18.005956] RSP: 002b:00007fa90affd8c0 EFLAGS: 00000202 [ 18.005956] RAX: 0000000000000003 RBX: 00007fa8e3a10000 RCX: 00007ffe7bd62a39 [ 18.005957] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 [ 18.005957] RBP: 00007fa90affd8c0 R08: 0000000000000000 R09: 0000000000000000 [ 18.005957] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 [ 18.005958] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 [ 18.005959] ---[ end trace c120e55e5ad35ff0 ]---
