In commit efb78fa86e95 ("lib/test_meminit: allocate pages up to order
MAX_ORDER"), the loop for testing pages is set to "<= MAX_ORDER" which
causes crashes in systems when run. Fix this to "< MAX_ORDER" to fix
the test to work properly.
Fixes: efb78fa86e95 ("lib/test_meminit: allocate pages up to order MAX_ORDER")
Cc: Andrew Donnellan <[email protected]>
Cc: Alexander Potapenko <[email protected]>
Cc: Xiaoke Wang <[email protected]>
Cc: <[email protected]>
Cc: Andrew Morton <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
---
lib/test_meminit.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/test_meminit.c b/lib/test_meminit.c
index 0ae35223d773..85d8dd8e01dc 100644
--- a/lib/test_meminit.c
+++ b/lib/test_meminit.c
@@ -93,7 +93,7 @@ static int __init test_pages(int *total_failures)
int failures = 0, num_tests = 0;
int i;
- for (i = 0; i <= MAX_ORDER; i++)
+ for (i = 0; i < MAX_ORDER; i++)
num_tests += do_alloc_pages_order(i, &failures);
REPORT_FAILURES_IN_FN();
--
2.42.0
On Thu, Oct 12, 2023 at 10:17 AM Greg Kroah-Hartman
<[email protected]> wrote:
>
> In commit efb78fa86e95 ("lib/test_meminit: allocate pages up to order
> MAX_ORDER"), the loop for testing pages is set to "<= MAX_ORDER" which
> causes crashes in systems when run. Fix this to "< MAX_ORDER" to fix
> the test to work properly.
What are the crashes you are seeing? Are those OOMs?
IIUC it should be valid to allocate with MAX_ORDER.
On Thu, Oct 12, 2023 at 10:40:14AM +0200, Alexander Potapenko wrote:
> On Thu, Oct 12, 2023 at 10:17 AM Greg Kroah-Hartman
> <[email protected]> wrote:
> >
> > In commit efb78fa86e95 ("lib/test_meminit: allocate pages up to order
> > MAX_ORDER"), the loop for testing pages is set to "<= MAX_ORDER" which
> > causes crashes in systems when run. Fix this to "< MAX_ORDER" to fix
> > the test to work properly.
>
> What are the crashes you are seeing? Are those OOMs?
They are WARN_ON() triggers. They are burried in the Android build
system, let me see if I can uncover them.
> IIUC it should be valid to allocate with MAX_ORDER.
"should", but I no longer get runtime warnings with this patch applied,
so something is wrong :)
Let me go dig for the logs again...
On Thu, Oct 12, 2023 at 12:26:58PM +0200, Greg Kroah-Hartman wrote:
> On Thu, Oct 12, 2023 at 10:40:14AM +0200, Alexander Potapenko wrote:
> > On Thu, Oct 12, 2023 at 10:17 AM Greg Kroah-Hartman
> > <[email protected]> wrote:
> > >
> > > In commit efb78fa86e95 ("lib/test_meminit: allocate pages up to order
> > > MAX_ORDER"), the loop for testing pages is set to "<= MAX_ORDER" which
> > > causes crashes in systems when run. Fix this to "< MAX_ORDER" to fix
> > > the test to work properly.
> >
> > What are the crashes you are seeing? Are those OOMs?
>
> They are WARN_ON() triggers. They are burried in the Android build
> system, let me see if I can uncover them.
>
> > IIUC it should be valid to allocate with MAX_ORDER.
>
> "should", but I no longer get runtime warnings with this patch applied,
> so something is wrong :)
>
> Let me go dig for the logs again...
Here's one, on the 5.4.y tree where this was backported to, with this
patch applied, it does not show up anymore:
[ 8.897682] test_stackinit: all tests passed!
[ 8.898200] init: Loaded kernel module /lib/modules/5.4.257-android11-2-00544-g5000f4d84d4f/kernel/lib/test_stackinit.ko
[ 8.898649] init: Loading module /lib/modules/5.4.257-android11-2-00544-g5000f4d84d4f/kernel/lib/test_meminit.ko with args ""
[ 9.015314] ------------[ cut here ]------------
[ 9.016359] WARNING: CPU: 0 PID: 1 at mm/page_alloc.c:4873 __alloc_pages_nodemask+0x1ad/0x310
[ 9.016946] Modules linked in: test_meminit(+) test_stackinit vmw_vsock_virtio_transport vmw_vsock_virtio_transport_common vsock_diag vsock snd_intel8x0 snd_ac97_codec ac97_bus gnss_cmdline gnss_serial dummy_cpufreq hci_vhci rtc_test virtio_net net_failover failover virt_wifi mac80211_hwsim virtio_pmem nd_virtio virtio_blk tpm_vtpm_proxy tpm virtio_console virtio_rng virtio_input virtio_pci virtio_mmio incrementalfs
[ 9.018268] CPU: 0 PID: 1 Comm: init Not tainted 5.4.257-android11-2-00544-g5000f4d84d4f-abP63746429 #1
[ 9.018707] Hardware name: ChromiumOS crosvm, BIOS 0
[ 9.018956] RIP: 0010:__alloc_pages_nodemask+0x1ad/0x310
[ 9.019178] Code: 89 f7 a9 00 00 0c 10 75 2b c6 45 c8 00 4c 39 65 b0 75 3c 48 8d 55 a8 89 7d a4 44 89 fe e8 eb 05 00 00 49 89 c5 e9 71 ff ff ff <0f> 0b 45 31 ed e9 74 ff ff ff 44 89 f1 a9 00 00 08 00 75 17 81 e1
[ 9.019922] RSP: 0018:ffffaf3ec000ba28 EFLAGS: 00010246
[ 9.020139] RAX: 86b5c88349dacd00 RBX: 0000000000000000 RCX: 0000000000000000
[ 9.020496] RDX: 0000000000000000 RSI: 000000000000000b RDI: 0000000000000cc0
[ 9.020843] RBP: ffffaf3ec000ba90 R08: ffffe222c8830000 R09: 0000000000000001
[ 9.021184] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
[ 9.021502] R13: 0000000000800000 R14: 0000000000000cc0 R15: ffffaf3ec000baec
[ 9.021829] FS: 00007f10e8024ff8(0000) GS:ffff8a5567a00000(0000) knlGS:0000000000000000
[ 9.022136] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 9.022295] CR2: 000055f8cc69cfd0 CR3: 0000000223ce0003 CR4: 0000000000360eb0
[ 9.022509] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 9.022890] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 9.023223] Call Trace:
[ 9.023473] ? show_regs+0x55/0x60
[ 9.023731] ? __warn+0xcb/0x150
[ 9.023896] ? __alloc_pages_nodemask+0x1ad/0x310
[ 9.024211] ? report_bug+0x11e/0x190
[ 9.024376] ? __alloc_pages_nodemask+0x1ad/0x310
[ 9.024598] ? do_invalid_op+0x2f/0xb0
[ 9.024823] ? invalid_op+0x1e/0x30
[ 9.024989] ? __alloc_pages_nodemask+0x1ad/0x310
[ 9.025213] do_alloc_pages_order+0x33/0xd2 [test_meminit]
[ 9.025429] ? __cfi_check+0x60/0x60 [test_meminit]
[ 9.025655] test_pages+0x34/0x9c [test_meminit]
[ 9.025869] init_module+0x2b/0x9a [test_meminit]
[ 9.026127] do_one_initcall+0x12e/0x390
[ 9.026363] do_init_module+0x4a/0x200
[ 9.026528] load_module+0x40b5/0x4560
[ 9.026781] ? kernel_read_file+0x18f/0x1e0
[ 9.026958] ? kernel_read_file_from_fd+0x52/0x80
[ 9.027143] __x64_sys_finit_module+0xcf/0x100
[ 9.027423] ? __x64_sys_rmdir.cfi_jt+0x8/0x8
[ 9.027642] do_syscall_64+0x6d/0xb0
[ 9.027802] entry_SYSCALL_64_after_hwframe+0x5c/0xc1
[ 9.028020] RIP: 0033:0x7f10e789a508
[ 9.028181] Code: c0 4c 89 1c 24 c3 e8 97 6c ff ff cc cc cc cc cc cc cc 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 72 09 f7 d8 89 c7 e8 77 55 05 00 c3 cc cc cc cc
[ 9.028913] RSP: 002b:00007ffd1d1f4a38 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
[ 9.029233] RAX: ffffffffffffffda RBX: 00007ffd1d1f4a71 RCX: 00007f10e789a508
[ 9.029547] RDX: 0000000000000000 RSI: 00007ffd1d1f4a71 RDI: 0000000000000006
[ 9.029880] RBP: 00007f10e82917f0 R08: 00007ffd1d1f4420 R09: 0000000000000050
[ 9.030202] R10: 0000000000000010 R11: 0000000000000246 R12: 00007f0ef721d010
[ 9.030518] R13: 00007ffd1d1f4a50 R14: 00007ffd1d1f4d50 R15: 0000000000000002
[ 9.030836] ---[ end trace afa5c86c1c294dbb ]---
[ 9.031252] general protection fault: 0000 [#1] PREEMPT SMP PTI
[ 9.031749] CPU: 0 PID: 1 Comm: init Tainted: G W 5.4.257-android11-2-00544-g5000f4d84d4f-abP63746429 #1
[ 9.032074] Hardware name: ChromiumOS crosvm, BIOS 0
[ 9.032229] RIP: 0010:fill_with_garbage_skip+0x2d/0x5e [test_meminit]
[ 9.032463] Code: e5 48 63 c6 48 39 d0 72 4e 48 01 d7 89 f0 29 d0 83 f8 04 72 29 29 d6 83 c6 fc c1 ee 02 48 8d 14 b5 04 00 00 00 48 ff c6 31 c9 <c7> 04 0f 9e ba a7 09 48 83 c1 04 48 39 ca 75 f0 29 c8 eb 02 31 f6
[ 9.033114] RSP: 0018:ffffaf3ec000ba90 EFLAGS: 00010246
[ 9.033376] RAX: 0000000000800000 RBX: 0000000000000000 RCX: 0000000000000000
[ 9.033754] RDX: 0000000000800000 RSI: 0000000000200000 RDI: 000701a340000000
[ 9.034104] RBP: ffffaf3ec000ba90 R08: ffffe222c8830000 R09: 0000000000000001
[ 9.034421] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
[ 9.034752] R13: 0000000000800000 R14: 000000000000000b R15: ffffaf3ec000baec
[ 9.035189] FS: 00007f10e8024ff8(0000) GS:ffff8a5567a00000(0000) knlGS:0000000000000000
[ 9.035563] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 9.035837] CR2: 000055f8cc69cfd0 CR3: 0000000223ce0003 CR4: 0000000000360eb0
[ 9.036191] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 9.036541] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 9.036859] Call Trace:
[ 9.037000] ? __die+0xdb/0x120
[ 9.037114] ? die+0x2a/0x50
[ 9.037309] ? do_general_protection+0x17c/0x1f0
[ 9.037530] ? general_protection+0x28/0x30
[ 9.037698] ? fill_with_garbage_skip+0x2d/0x5e [test_meminit]
[ 9.038139] do_alloc_pages_order+0x55/0xd2 [test_meminit]
[ 9.038340] ? __cfi_check+0x60/0x60 [test_meminit]
[ 9.038614] test_pages+0x34/0x9c [test_meminit]
[ 9.038835] init_module+0x2b/0x9a [test_meminit]
[ 9.039047] do_one_initcall+0x12e/0x390
[ 9.039185] do_init_module+0x4a/0x200
[ 9.039352] load_module+0x40b5/0x4560
[ 9.039523] ? kernel_read_file+0x18f/0x1e0
[ 9.039703] ? kernel_read_file_from_fd+0x52/0x80
[ 9.039910] __x64_sys_finit_module+0xcf/0x100
[ 9.040130] ? __x64_sys_rmdir.cfi_jt+0x8/0x8
[ 9.040262] do_syscall_64+0x6d/0xb0
[ 9.040437] entry_SYSCALL_64_after_hwframe+0x5c/0xc1
[ 9.040626] RIP: 0033:0x7f10e789a508
[ 9.040779] Code: c0 4c 89 1c 24 c3 e8 97 6c ff ff cc cc cc cc cc cc cc 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 72 09 f7 d8 89 c7 e8 77 55 05 00 c3 cc cc cc cc
[ 9.041509] RSP: 002b:00007ffd1d1f4a38 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
[ 9.041965] RAX: ffffffffffffffda RBX: 00007ffd1d1f4a71 RCX: 00007f10e789a508
[ 9.042362] RDX: 0000000000000000 RSI: 00007ffd1d1f4a71 RDI: 0000000000000006
[ 9.042724] RBP: 00007f10e82917f0 R08: 00007ffd1d1f4420 R09: 0000000000000050
[ 9.043055] R10: 0000000000000010 R11: 0000000000000246 R12: 00007f0ef721d010
[ 9.043379] R13: 00007ffd1d1f4a50 R14: 00007ffd1d1f4d50 R15: 0000000000000002
[ 9.043977] Modules linked in: test_meminit(+) test_stackinit vmw_vsock_virtio_transport vmw_vsock_virtio_transport_common vsock_diag vsock snd_intel8x0 snd_ac97_codec ac97_bus gnss_cmdline gnss_serial dummy_cpufreq hci_vhci rtc_test virtio_net net_failover failover virt_wifi mac80211_hwsim virtio_pmem nd_virtio virtio_blk tpm_vtpm_proxy tpm virtio_console virtio_rng virtio_input virtio_pci virtio_mmio incrementalfs
[ 9.045829] ---[ end trace afa5c86c1c294dbc ]---
[ 9.046415] RIP: 0010:fill_with_garbage_skip+0x2d/0x5e [test_meminit]
[ 9.046706] Code: e5 48 63 c6 48 39 d0 72 4e 48 01 d7 89 f0 29 d0 83 f8 04 72 29 29 d6 83 c6 fc c1 ee 02 48 8d 14 b5 04 00 00 00 48 ff c6 31 c9 <c7> 04 0f 9e ba a7 09 48 83 c1 04 48 39 ca 75 f0 29 c8 eb 02 31 f6
[ 9.047388] RSP: 0018:ffffaf3ec000ba90 EFLAGS: 00010246
[ 9.047552] RAX: 0000000000800000 RBX: 0000000000000000 RCX: 0000000000000000
[ 9.047798] RDX: 0000000000800000 RSI: 0000000000200000 RDI: 000701a340000000
[ 9.048048] RBP: ffffaf3ec000ba90 R08: ffffe222c8830000 R09: 0000000000000001
[ 9.048291] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
[ 9.048534] R13: 0000000000800000 R14: 000000000000000b R15: ffffaf3ec000baec
[ 9.048780] FS: 00007f10e8024ff8(0000) GS:ffff8a5567a00000(0000) knlGS:0000000000000000
[ 9.049025] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 9.049336] CR2: 000055f8cc69cfd0 CR3: 0000000223ce0003 CR4: 0000000000360eb0
[ 9.049706] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 9.050054] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 9.050387] Kernel panic - not syncing: Fatal exception
[ 9.052413] Kernel Offset: 0x2b000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
On Thu, Oct 12, 2023 at 12:37 PM Greg Kroah-Hartman
<[email protected]> wrote:
>
> On Thu, Oct 12, 2023 at 12:26:58PM +0200, Greg Kroah-Hartman wrote:
> > On Thu, Oct 12, 2023 at 10:40:14AM +0200, Alexander Potapenko wrote:
> > > On Thu, Oct 12, 2023 at 10:17 AM Greg Kroah-Hartman
> > > <[email protected]> wrote:
> > > >
> > > > In commit efb78fa86e95 ("lib/test_meminit: allocate pages up to order
> > > > MAX_ORDER"), the loop for testing pages is set to "<= MAX_ORDER" which
> > > > causes crashes in systems when run. Fix this to "< MAX_ORDER" to fix
> > > > the test to work properly.
> > >
> > > What are the crashes you are seeing? Are those OOMs?
> >
> > They are WARN_ON() triggers. They are burried in the Android build
> > system, let me see if I can uncover them.
> >
> > > IIUC it should be valid to allocate with MAX_ORDER.
> >
> > "should", but I no longer get runtime warnings with this patch applied,
> > so something is wrong :)
I think I know what's going on. In March 2023 Kirill changed the
semantics of MAX_ORDER to be inclusive: now alloc_pages() accepts
0..MAX_ORDER, whereas previously it was 0..MAX_ORDER-1:
https://github.com/torvalds/linux/commit/23baf831a32c04f9a968812511540b1b3e648bf5
Older kernel versions had an explicit check for order >= MAX_ORDER,
which is what you're seeing on Android.
On Thu, Oct 12, 2023 at 01:14:39PM +0200, Alexander Potapenko wrote:
> On Thu, Oct 12, 2023 at 12:37 PM Greg Kroah-Hartman
> <[email protected]> wrote:
> >
> > On Thu, Oct 12, 2023 at 12:26:58PM +0200, Greg Kroah-Hartman wrote:
> > > On Thu, Oct 12, 2023 at 10:40:14AM +0200, Alexander Potapenko wrote:
> > > > On Thu, Oct 12, 2023 at 10:17 AM Greg Kroah-Hartman
> > > > <[email protected]> wrote:
> > > > >
> > > > > In commit efb78fa86e95 ("lib/test_meminit: allocate pages up to order
> > > > > MAX_ORDER"), the loop for testing pages is set to "<= MAX_ORDER" which
> > > > > causes crashes in systems when run. Fix this to "< MAX_ORDER" to fix
> > > > > the test to work properly.
> > > >
> > > > What are the crashes you are seeing? Are those OOMs?
> > >
> > > They are WARN_ON() triggers. They are burried in the Android build
> > > system, let me see if I can uncover them.
> > >
> > > > IIUC it should be valid to allocate with MAX_ORDER.
> > >
> > > "should", but I no longer get runtime warnings with this patch applied,
> > > so something is wrong :)
>
> I think I know what's going on. In March 2023 Kirill changed the
> semantics of MAX_ORDER to be inclusive: now alloc_pages() accepts
> 0..MAX_ORDER, whereas previously it was 0..MAX_ORDER-1:
> https://github.com/torvalds/linux/commit/23baf831a32c04f9a968812511540b1b3e648bf5
> Older kernel versions had an explicit check for order >= MAX_ORDER,
> which is what you're seeing on Android.
Thanks for digging this up. I'll just queue this patch up for 6.1 and
older stable kernels then as the commit above you found is in 6.4 so
nothing newer than that will need this fix.
Thanks for the quick response!
greg k-h