Message ID | 1442431574-4575-1-git-send-email-kda@linux-powerpc.org (mailing list archive) |
---|---|
State | Accepted |
Headers | show |
On Wed, Sep 16, 2015 at 10:26:14PM +0300, Denis Kirjanov wrote: > During the MSI bitmap test on boot kmemleak spews the following trace: > > unreferenced object 0xc00000016e86c900 (size 64): > comm "swapper/0", pid 1, jiffies 4294893173 (age 518.024s) > hex dump (first 32 bytes): > 00 00 01 ff 7f ff 7f 37 00 00 00 00 00 00 00 00 > .......7........ > ff ff ff ff ff ff ff ff 01 ff ff ff ff > ff ff ff > ................ > backtrace: > [<c00000000003eebc>] .zalloc_maybe_bootmem+0x3c/0x380 > [<c000000000042d6c>] .msi_bitmap_alloc+0x3c/0xb0 > [<c000000000a9aff8>] .msi_bitmap_selftest+0x30/0x2b4 > [<c0000000000090f4>] .do_one_initcall+0xd4/0x270 > [<c000000000a8e250>] .kernel_init_freeable+0x1a0/0x280 > [<c000000000009b5c>] .kernel_init+0x1c/0x120 > [<c000000000007fbc>] .ret_from_kernel_thread+0x58/0x9c > > Added a flag to msi_bitmap for tracking allocations > from slab and memblock so we can properly free/handle > memory in msi_bitmap_free(). > > Signed-off-by: Denis Kirjanov <kda@linux-powerpc.org> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
On 9/17/15, Catalin Marinas <catalin.marinas@arm.com> wrote: > On Wed, Sep 16, 2015 at 10:26:14PM +0300, Denis Kirjanov wrote: >> During the MSI bitmap test on boot kmemleak spews the following trace: >> >> unreferenced object 0xc00000016e86c900 (size 64): >> comm "swapper/0", pid 1, jiffies 4294893173 (age 518.024s) >> hex dump (first 32 bytes): >> 00 00 01 ff 7f ff 7f 37 00 00 00 00 00 00 00 00 >> .......7........ >> ff ff ff ff ff ff ff ff 01 ff ff ff ff >> ff ff ff >> ................ >> backtrace: >> [<c00000000003eebc>] .zalloc_maybe_bootmem+0x3c/0x380 >> [<c000000000042d6c>] .msi_bitmap_alloc+0x3c/0xb0 >> [<c000000000a9aff8>] .msi_bitmap_selftest+0x30/0x2b4 >> [<c0000000000090f4>] .do_one_initcall+0xd4/0x270 >> [<c000000000a8e250>] .kernel_init_freeable+0x1a0/0x280 >> [<c000000000009b5c>] .kernel_init+0x1c/0x120 >> [<c000000000007fbc>] .ret_from_kernel_thread+0x58/0x9c >> >> Added a flag to msi_bitmap for tracking allocations >> from slab and memblock so we can properly free/handle >> memory in msi_bitmap_free(). >> >> Signed-off-by: Denis Kirjanov <kda@linux-powerpc.org> > > Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> > Hi Michael, could you please apply the patch to fixes branch as well? Thanks!
On Wed, 2015-16-09 at 19:26:14 UTC, Denis Kirjanov wrote: > During the MSI bitmap test on boot kmemleak spews the following trace: > > unreferenced object 0xc00000016e86c900 (size 64): > comm "swapper/0", pid 1, jiffies 4294893173 (age 518.024s) > hex dump (first 32 bytes): > 00 00 01 ff 7f ff 7f 37 00 00 00 00 00 00 00 00 > .......7........ > ff ff ff ff ff ff ff ff 01 ff ff ff ff > ff ff ff > ................ > backtrace: > [<c00000000003eebc>] .zalloc_maybe_bootmem+0x3c/0x380 > [<c000000000042d6c>] .msi_bitmap_alloc+0x3c/0xb0 > [<c000000000a9aff8>] .msi_bitmap_selftest+0x30/0x2b4 > [<c0000000000090f4>] .do_one_initcall+0xd4/0x270 > [<c000000000a8e250>] .kernel_init_freeable+0x1a0/0x280 > [<c000000000009b5c>] .kernel_init+0x1c/0x120 > [<c000000000007fbc>] .ret_from_kernel_thread+0x58/0x9c > > Added a flag to msi_bitmap for tracking allocations > from slab and memblock so we can properly free/handle > memory in msi_bitmap_free(). > > Signed-off-by: Denis Kirjanov <kda@linux-powerpc.org> Applied to powerpc next, thanks. https://git.kernel.org/powerpc/c/cb2d3883c6033831e2a93b39 cheers
diff --git a/arch/powerpc/include/asm/msi_bitmap.h b/arch/powerpc/include/asm/msi_bitmap.h index 97ac3f4..47a8ebc 100644 --- a/arch/powerpc/include/asm/msi_bitmap.h +++ b/arch/powerpc/include/asm/msi_bitmap.h @@ -19,6 +19,7 @@ struct msi_bitmap { unsigned long *bitmap; spinlock_t lock; unsigned int irq_count; + bool bitmap_from_slab; }; int msi_bitmap_alloc_hwirqs(struct msi_bitmap *bmp, int num); diff --git a/arch/powerpc/sysdev/msi_bitmap.c b/arch/powerpc/sysdev/msi_bitmap.c index 73b64c7..6f2f3d6 100644 --- a/arch/powerpc/sysdev/msi_bitmap.c +++ b/arch/powerpc/sysdev/msi_bitmap.c @@ -11,6 +11,7 @@ #include <linux/slab.h> #include <linux/kernel.h> #include <linux/bitmap.h> +#include <linux/bootmem.h> #include <asm/msi_bitmap.h> #include <asm/setup.h> @@ -122,7 +123,16 @@ int msi_bitmap_alloc(struct msi_bitmap *bmp, unsigned int irq_count, size = BITS_TO_LONGS(irq_count) * sizeof(long); pr_debug("msi_bitmap: allocator bitmap size is 0x%x bytes\n", size); - bmp->bitmap = zalloc_maybe_bootmem(size, GFP_KERNEL); + if (slab_is_available()) { + bmp->bitmap = kzalloc(size, GFP_KERNEL); + bmp->bitmap_from_slab = true; + } else { + bmp->bitmap = memblock_virt_alloc(size, 0); + bmp->bitmap_from_slab = false; + /* the bitmap won't be freed from memblock allocator */ + kmemleak_not_leak(bmp->bitmap); + } + if (!bmp->bitmap) { pr_debug("msi_bitmap: ENOMEM allocating allocator bitmap!\n"); return -ENOMEM; @@ -138,7 +148,8 @@ int msi_bitmap_alloc(struct msi_bitmap *bmp, unsigned int irq_count, void msi_bitmap_free(struct msi_bitmap *bmp) { - /* we can't free the bitmap we don't know if it's bootmem etc. */ + if (bmp->bitmap_from_slab) + kfree(bmp->bitmap); of_node_put(bmp->of_node); bmp->bitmap = NULL; } @@ -203,8 +214,6 @@ static void __init test_basics(void) /* Clients may WARN_ON bitmap == NULL for "not-allocated" */ WARN_ON(bmp.bitmap != NULL); - - kfree(bmp.bitmap); } static void __init test_of_node(void)
During the MSI bitmap test on boot kmemleak spews the following trace: unreferenced object 0xc00000016e86c900 (size 64): comm "swapper/0", pid 1, jiffies 4294893173 (age 518.024s) hex dump (first 32 bytes): 00 00 01 ff 7f ff 7f 37 00 00 00 00 00 00 00 00 .......7........ ff ff ff ff ff ff ff ff 01 ff ff ff ff ff ff ff ................ backtrace: [<c00000000003eebc>] .zalloc_maybe_bootmem+0x3c/0x380 [<c000000000042d6c>] .msi_bitmap_alloc+0x3c/0xb0 [<c000000000a9aff8>] .msi_bitmap_selftest+0x30/0x2b4 [<c0000000000090f4>] .do_one_initcall+0xd4/0x270 [<c000000000a8e250>] .kernel_init_freeable+0x1a0/0x280 [<c000000000009b5c>] .kernel_init+0x1c/0x120 [<c000000000007fbc>] .ret_from_kernel_thread+0x58/0x9c Added a flag to msi_bitmap for tracking allocations from slab and memblock so we can properly free/handle memory in msi_bitmap_free(). Signed-off-by: Denis Kirjanov <kda@linux-powerpc.org> Changes vor v2: - added a flag to msi_bitmap - kmemleak annotaion moved under CONFIG_MSI_BITMAP_SELFTEST Changes vor v3: - correctly initialize bitmap_from_slab - put kmemleak_not_leak annotation right after the allocation --- arch/powerpc/include/asm/msi_bitmap.h | 1 + arch/powerpc/sysdev/msi_bitmap.c | 17 +++++++++++++---- 2 files changed, 14 insertions(+), 4 deletions(-)