[PATCH v4 0/3] Improve the dma-buf tracking

Currently, all dma-bufs share the same anonymous inode. While we can count how many dma-buf fds or mappings a process has, we can't get the size of the backing buffers or tell if two entries point to the same dma-buf. And in debugfs, we can get a per-buffer breakdown of size and reference count,

[PATCH v4 2/3] dma-buf: add DMA_BUF_SET_NAME ioctls

-off-by: Chenbo Feng --- drivers/dma-buf/dma-buf.c| 64 ++-- include/linux/dma-buf.h | 5 ++- include/uapi/linux/dma-buf.h | 3 ++ 3 files changed, 68 insertions(+), 4 deletions(-) diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c index

[PATCH v4 3/3] dma-buf: add show_fdinfo handler

From: Greg Hackmann The show_fdinfo handler exports the same information available through debugfs on a per-buffer basis. Signed-off-by: Greg Hackmann Signed-off-by: Chenbo Feng --- drivers/dma-buf/dma-buf.c | 15 +++ 1 file changed, 15 insertions(+) diff --git a/drivers/dma-buf

Re: [RESEND PATCH v3 2/3] dma-buf: add DMA_BUF_{GET,SET}_NAME ioctls

On Wed, Jun 12, 2019 at 7:43 AM Sumit Semwal wrote: > > Hello Chenbo, > > Thanks very much for your patches. Other than a couple tiny nits > below, I think these look good, and I will merge them before the end > of this week. > On Tue, 11 Jun 2019 at 05:32, Chenbo Feng wrot

[RESEND PATCH v3 1/3] dma-buf: give each buffer a full-fledged inode

for dma-buf, following the example in fs/aio.c. Signed-off-by: Greg Hackmann Signed-off-by: Chenbo Feng --- drivers/dma-buf/dma-buf.c | 63 ++ include/uapi/linux/magic.h | 1 + 2 files changed, 58 insertions(+), 6 deletions(-) diff --git a/drivers/dma-buf

[RESEND PATCH v3 1/3] dma-buf: give each buffer a full-fledged inode

for dma-buf, following the example in fs/aio.c. Signed-off-by: Greg Hackmann Signed-off-by: Chenbo Feng --- drivers/dma-buf/dma-buf.c | 63 ++ include/uapi/linux/magic.h | 1 + 2 files changed, 58 insertions(+), 6 deletions(-) diff --git a/drivers/dma-buf

[RFC v2 2/3] dma-buf: add DMA_BUF_{GET,SET}_NAME ioctls

-by: Greg Hackmann Signed-off-by: Chenbo Feng --- drivers/dma-buf/dma-buf.c| 48 ++-- include/linux/dma-buf.h | 5 +++- include/uapi/linux/dma-buf.h | 4 +++ 3 files changed, 54 insertions(+), 3 deletions(-) diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma

[RFC dma-buf 3/3] dma-buf: add show_fdinfo handler

From: Greg Hackmann The show_fdinfo handler exports the same information available through debugfs on a per-buffer basis. Signed-off-by: Greg Hackmann Signed-off-by: Chenbo Feng --- drivers/dma-buf/dma-buf.c | 15 +++ 1 file changed, 15 insertions(+) diff --git a/drivers/dma-buf

[RFC dma-buf 2/3] dma-buf: add DMA_BUF_{GET,SET}_NAME ioctls

-by: Greg Hackmann Signed-off-by: Chenbo Feng --- drivers/dma-buf/dma-buf.c| 42 ++-- include/linux/dma-buf.h | 5 - include/uapi/linux/dma-buf.h | 4 3 files changed, 48 insertions(+), 3 deletions(-) diff --git a/drivers/dma-buf/dma-buf.c b/drivers

[RFC dma-buf 1/3] dma-buf: give each buffer a full-fledged inode

for dma-buf, following the example in fs/aio.c. Signed-off-by: Greg Hackmann Signed-off-by: Chenbo Feng --- drivers/dma-buf/dma-buf.c | 68 +- include/uapi/linux/magic.h | 1 + 2 files changed, 61 insertions(+), 8 deletions(-) diff --git a/drivers/dma-buf

[RFC dma-buf 0/3] Improve the dma-buf tracking

Currently, all dma-bufs share the same anonymous inode. While we can count how many dma-buf fds or mappings a process has, we can't get the size of the backing buffers or tell if two entries point to the same dma-buf. And in debugfs, we can get a per-buffer breakdown of size and reference count,

Re: [PATCH AUTOSEL for 4.4 065/162] tcp: Add a tcp_filter hook before handle ack packet

On Sun, Apr 8, 2018 at 5:28 PM, Sasha Levin <alexander.le...@microsoft.com> wrote: > From: Chenbo Feng <fe...@google.com> > > [ Upstream commit 8fac365f63c866a00015fa13932d8ffc584518b8 ] > > Currently in both ipv4 and ipv6 code path, the ack packet received when >

Re: [PATCH AUTOSEL for 4.4 065/162] tcp: Add a tcp_filter hook before handle ack packet

On Sun, Apr 8, 2018 at 5:28 PM, Sasha Levin wrote: > From: Chenbo Feng > > [ Upstream commit 8fac365f63c866a00015fa13932d8ffc584518b8 ] > > Currently in both ipv4 and ipv6 code path, the ack packet received when > sk at TCP_NEW_SYN_RECV state is not filtered by socket filter