| Message ID | 20220818165906.64450-2-toke@redhat.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Delegated to: | BPF |
| Headers | show |
| Series | A couple of small refactorings of BPF program call sites | expand |
| Context | Check | Description |
|---|---|---|
| bpf/vmtest-bpf-next-PR | fail | PR summary |
| netdev/tree_selection | success | Clearly marked for bpf-next, async |
| netdev/fixes_present | success | Fixes tag not required for -next series |
| netdev/subject_prefix | success | Link |
| netdev/cover_letter | success | Series has a cover letter |
| netdev/patch_count | success | Link |
| netdev/header_inline | success | No static functions without inline keyword in header files |
| netdev/build_32bit | success | Errors and warnings before: 445444 this patch: 445444 |
| netdev/cc_maintainers | success | CCed 6 of 6 maintainers |
| netdev/build_clang | success | Errors and warnings before: 1067 this patch: 1067 |
| netdev/module_param | success | Was 0 now: 0 |
| netdev/verify_signedoff | success | Signed-off-by tag matches author and committer |
| netdev/check_selftest | success | No net selftest shell script |
| netdev/verify_fixes | success | No Fixes tag |
| netdev/build_allmodconfig_warn | success | Errors and warnings before: 466748 this patch: 466748 |
| netdev/checkpatch | success | total: 0 errors, 0 warnings, 0 checks, 14 lines checked |
| netdev/kdoc | success | Errors and warnings before: 0 this patch: 0 |
| netdev/source_inline | success | Was 0 now: 0 |
| bpf/vmtest-bpf-next-VM_Test-3 | fail | Logs for Kernel LATEST on z15 with gcc |
| bpf/vmtest-bpf-next-VM_Test-4 | success | Logs for llvm-toolchain |
| bpf/vmtest-bpf-next-VM_Test-5 | success | Logs for set-matrix |
| bpf/vmtest-bpf-next-VM_Test-2 | fail | Logs for Kernel LATEST on ubuntu-latest with llvm-16 |
| bpf/vmtest-bpf-next-VM_Test-1 | success | Logs for Kernel LATEST on ubuntu-latest with gcc |
On Thu, 18 Aug 2022 18:59:03 +0200 Toke Høiland-Jørgensen wrote: > Move the received_rps counter value next to the other RPS-related members > in softnet_data. This closes two four-byte holes in the structure, making > room for another pointer in the first two cache lines without bumping the > xmit struct to its own line. What's the pointer you're making space for (which I hope will explain why this patch is part of this otherwise bpf series)?
Jakub Kicinski <kuba@kernel.org> writes: > On Thu, 18 Aug 2022 18:59:03 +0200 Toke Høiland-Jørgensen wrote: >> Move the received_rps counter value next to the other RPS-related members >> in softnet_data. This closes two four-byte holes in the structure, making >> room for another pointer in the first two cache lines without bumping the >> xmit struct to its own line. > > What's the pointer you're making space for (which I hope will explain > why this patch is part of this otherwise bpf series)? The XDP queueing series adds a pointer to keep track of which interfaces were scheduled for transmission using the XDP dequeue hook (similar to how the qdisc wake code works): https://lore.kernel.org/r/20220713111430.134810-12-toke@redhat.com Note that it's still up in the air if this ends up being the way this will be implemented, so I'm OK with dropping this patch for now if you'd rather wait until it's really needed. OTOH it also seemed like a benign change on its own, so I figured I might as well include this patch when sending these out. WDYT? -Toke
On Fri, 19 Aug 2022 14:38:14 +0200 Toke Høiland-Jørgensen wrote: > Jakub Kicinski <kuba@kernel.org> writes: > > On Thu, 18 Aug 2022 18:59:03 +0200 Toke Høiland-Jørgensen wrote: > >> Move the received_rps counter value next to the other RPS-related members > >> in softnet_data. This closes two four-byte holes in the structure, making > >> room for another pointer in the first two cache lines without bumping the > >> xmit struct to its own line. > > > > What's the pointer you're making space for (which I hope will explain > > why this patch is part of this otherwise bpf series)? > > The XDP queueing series adds a pointer to keep track of which interfaces > were scheduled for transmission using the XDP dequeue hook (similar to > how the qdisc wake code works): > > https://lore.kernel.org/r/20220713111430.134810-12-toke@redhat.com I see, it makes more sense now :) > Note that it's still up in the air if this ends up being the way this > will be implemented, so I'm OK with dropping this patch for now if you'd > rather wait until it's really needed. OTOH it also seemed like a benign > change on its own, so I figured I might as well include this patch when > sending these out. WDYT? Whatever is easiest :)
Jakub Kicinski <kuba@kernel.org> writes: > On Fri, 19 Aug 2022 14:38:14 +0200 Toke Høiland-Jørgensen wrote: >> Jakub Kicinski <kuba@kernel.org> writes: >> > On Thu, 18 Aug 2022 18:59:03 +0200 Toke Høiland-Jørgensen wrote: >> >> Move the received_rps counter value next to the other RPS-related members >> >> in softnet_data. This closes two four-byte holes in the structure, making >> >> room for another pointer in the first two cache lines without bumping the >> >> xmit struct to its own line. >> > >> > What's the pointer you're making space for (which I hope will explain >> > why this patch is part of this otherwise bpf series)? >> >> The XDP queueing series adds a pointer to keep track of which interfaces >> were scheduled for transmission using the XDP dequeue hook (similar to >> how the qdisc wake code works): >> >> https://lore.kernel.org/r/20220713111430.134810-12-toke@redhat.com > > I see, it makes more sense now :) > >> Note that it's still up in the air if this ends up being the way this >> will be implemented, so I'm OK with dropping this patch for now if you'd >> rather wait until it's really needed. OTOH it also seemed like a benign >> change on its own, so I figured I might as well include this patch when >> sending these out. WDYT? > > Whatever is easiest :) Alright, I'm OK with either so let's leave it up to the (BPF) maintainers if they want to drop this patch or just merge the whole series? :) -Toke
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 1a3cb93c3dcc..fe9aeca2fce9 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3100,7 +3100,6 @@ struct softnet_data { /* stats */ unsigned int processed; unsigned int time_squeeze; - unsigned int received_rps; #ifdef CONFIG_RPS struct softnet_data *rps_ipi_list; #endif @@ -3133,6 +3132,7 @@ struct softnet_data { unsigned int cpu; unsigned int input_queue_tail; #endif + unsigned int received_rps; unsigned int dropped; struct sk_buff_head input_pkt_queue; struct napi_struct backlog;
Move the received_rps counter value next to the other RPS-related members in softnet_data. This closes two four-byte holes in the structure, making room for another pointer in the first two cache lines without bumping the xmit struct to its own line. Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com> --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)