Re: [PATCH v5 00/18] IOMMUFD Dirty Tracking

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, Oct 23, 2023 at 07:21:09PM +0100, Joao Martins wrote:
> External email: Use caution opening links or attachments
> 
> 
> On 23/10/2023 19:12, Nicolin Chen wrote:
> > On Mon, Oct 23, 2023 at 12:49:55PM +0100, Joao Martins wrote:
> >> Here's an example down that avoids the kernel header dependency; imported from
> >> the arch-independent non-atomic bitops
> >> (include/asm-generic/bitops/generic-non-atomic.h)
> >>
> >> diff --git a/tools/testing/selftests/iommu/iommufd.c
> >> b/tools/testing/selftests/iommu/iommufd.c
> >> index 96837369a0aa..026ff9f5c1f3 100644
> >> --- a/tools/testing/selftests/iommu/iommufd.c
> >> +++ b/tools/testing/selftests/iommu/iommufd.c
> >> @@ -12,7 +12,6 @@
> >>  static unsigned long HUGEPAGE_SIZE;
> >>
> >>  #define MOCK_PAGE_SIZE (PAGE_SIZE / 2)
> >> -#define BITS_PER_BYTE 8
> >>
> >>  static unsigned long get_huge_page_size(void)
> >>  {
> >> diff --git a/tools/testing/selftests/iommu/iommufd_utils.h
> >> b/tools/testing/selftests/iommu/iommufd_utils.h
> >> index 390563ff7935..6bbcab7fd6ab 100644
> >> --- a/tools/testing/selftests/iommu/iommufd_utils.h
> >> +++ b/tools/testing/selftests/iommu/iommufd_utils.h
> >> @@ -9,8 +9,6 @@
> >>  #include <sys/ioctl.h>
> >>  #include <stdint.h>
> >>  #include <assert.h>
> >> -#include <linux/bitmap.h>
> >> -#include <linux/bitops.h>
> >>
> >>  #include "../kselftest_harness.h"
> >>  #include "../../../../drivers/iommu/iommufd/iommufd_test.h"
> >> @@ -18,6 +16,24 @@
> >>  /* Hack to make assertions more readable */
> >>  #define _IOMMU_TEST_CMD(x) IOMMU_TEST_CMD
> >>
> >> +/* Imported from include/asm-generic/bitops/generic-non-atomic.h */
> >> +#define BITS_PER_BYTE 8
> >> +#define BITS_PER_LONG __BITS_PER_LONG
> >> +#define BIT_MASK(nr) (1UL << ((nr) % __BITS_PER_LONG))
> >> +#define BIT_WORD(nr) ((nr) / __BITS_PER_LONG)
> >> +
> >> +static inline void set_bit(unsigned int nr, unsigned long *addr)
> >
> > The whole piece could fix the break, except this one. We'd need
> > __set_bit instead of set_bit.
> >
> 
> I changed it set_bit in the caller of course

Can you confirm the test results too? I am seeing test failing
and BUG_ON since this commit:

1d2ac3b64486 (HEAD) iommufd/selftest: Test out_capabilities in IOMMU_GET_HW_INFO

-----logs-----
# ok 133 iommufd_dirty_tracking.domain_dirty128k.set_dirty_tracking
# #  RUN           iommufd_dirty_tracking.domain_dirty128k.device_dirty_capability ... 
# # iommufd.c:1577:device_dirty_capability:Expected IOMMU_HW_CAP_DIRTY_TRACKING (1) == caps & IOMMU_HW_CAP_DIRTY_TRACKING (0)
# # device_dirty_capability: Test terminated by assertion
.....
# # FAILED: 151 / 161 tests passed.

-----bug_on-----
[   29.209521] BUG: unable to handle page fault for address: 000056258adc0000
[   29.209771] #PF: supervisor read access in kernel mode
[   29.209965] #PF: error_code(0x0001) - permissions violation
[   29.210155] PGD 112975067 P4D 112975067 PUD 112976067 PMD 10e5a8067 PTE 800000010973b067
[   29.210446] Oops: 0001 [#1] SMP
[   29.210594] CPU: 1 PID: 857 Comm: iommufd Not tainted 6.6.0-rc2+ #1823
[   29.210842] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
[   29.211267] RIP: 0010:iommufd_test+0xb7a/0x1120 [iommufd]
[   29.211480] Code: 82 a8 00 00 00 4c 8b 7d 90 c7 45 98 00 00 00 00 4d 89 cd 4c 89 5d 88 4c 89 75 80 48 89 45 a8 4c 89 95 78 ff ff ff 48 8b 45 b8 <48> 0f a3 18 73 5e 48 8b 4d a8 31 d2 4c 89 e8 49 f7 f4 48 89 c6 48
[   29.212131] RSP: 0018:ffffc900029f7d70 EFLAGS: 00010206
[   29.212348] RAX: 000056258adc0000 RBX: 0000000000000000 RCX: ffff888104296498
[   29.212638] RDX: 0000000000000000 RSI: 0000000094904f49 RDI: ffff888103fa23c8
[   29.212928] RBP: ffffc900029f7e00 R08: 0000000000020000 R09: 0000000001000000
[   29.213214] R10: ffffc900029f7e10 R11: ffffc900029f7e30 R12: 0000000000000800
[   29.213501] R13: 0000000001000000 R14: ffff888104296400 R15: 0000000000000040
[   29.213786] FS:  00007f485e907740(0000) GS:ffff8881ba440000(0000) knlGS:0000000000000000
[   29.214072] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   29.214310] CR2: 000056258adc0000 CR3: 0000000112974001 CR4: 00000000003706a0
[   29.214592] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   29.214882] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[   29.215171] Call Trace:
[   29.215265]  <TASK>
[   29.215358]  ? show_regs+0x5c/0x70
[   29.215496]  ? __die+0x1f/0x60
[   29.215638]  ? page_fault_oops+0x15d/0x440
[   29.215779]  ? exc_page_fault+0x4ca/0x9e0
[   29.215923]  ? lock_acquire+0xb8/0x2a0
[   29.216064]  ? asm_exc_page_fault+0x27/0x30
[   29.216207]  ? iommufd_test+0xb7a/0x1120 [iommufd]
[   29.216435]  ? should_fail_usercopy+0x15/0x20
[   29.216577]  iommufd_fops_ioctl+0x10d/0x190 [iommufd]
[   29.216725]  __x64_sys_ioctl+0x412/0x9b0
[   29.216823]  do_syscall_64+0x3c/0x80
[   29.216919]  entry_SYSCALL_64_after_hwframe+0x46/0xb0
[   29.217050] RIP: 0033:0x7f485ea0d04f
[   29.217147] Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 18 48 8b 44 24 18 64 48 2b 04 25 28 00 00
[   29.217593] RSP: 002b:00007ffddef77ca0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[   29.217785] RAX: ffffffffffffffda RBX: 0000562588e29d60 RCX: 00007f485ea0d04f
[   29.217982] RDX: 00007ffddef77d30 RSI: 0000000000003ba0 RDI: 0000000000000005
[   29.218177] RBP: 0000562588e29038 R08: 4000000000000000 R09: 0000000000000008
[   29.218372] R10: 0000000000000001 R11: 0000000000000246 R12: 000056258adc0000
[   29.218567] R13: 0000000000000000 R14: 0000000000000040 R15: 0000562588e29d60
[   29.218772]  </TASK>
[   29.218835] Modules linked in: iommufd ib_umad rdma_ucm rdma_cm ib_ipoib iw_cm ib_cm mlx5_ib ib_uverbs ib_core mlx5_core
[   29.219117] CR2: 000056258adc0000
[   29.219216] ---[ end trace 0000000000000000 ]---





[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux