similar to: timing related vunlerability that reveals whether files exist without regard to permissions

Package: xen-hypervisor-4.0-amd64 Version: 4.0.1-1 Severity: normal My 32Bit Dom0-crashes with error "Out of SW-IOMMU space for 65536 bytes" after doing moderate I/O on the local storage, which is a software-raid1 over 3 harddrives, connected to the builtin Perc H200 controller with ext3-fs. The error is reproduceable with "dd if=/dev/zero of=bigfile bs=1024 count=1000000"

Previously we measured the total time and used it to calculate the rate of different operations. This is incorrect and hides the real throughput. A more useful way is to measure the time we spent in each operation. Here is an example run with this change: $ ./nbdkit --foreground \ --unix /tmp/nbd.sock \ --exportname '' \ --filter stats \ file file=/var/tmp/dst.img \

I find bytes and bits-per-second unhelpful and hard to parse. Using GiB for sizes works for common disk images, and MiB/s works for common storage throughput. Here is an example run with this change: $ ./nbdkit --foreground \ --unix /tmp/nbd.sock \ --exportname '' \ --filter stats \ file file=/var/tmp/dst.img \ statsfile=/dev/stderr \ --run 'qemu-img convert

I find bytes and bits-per-second unhelpful and hard to parse. Add also size in GiB, and show rate in MiB per second. This works well for common disk images and storage. Here is an example run with this change: $ ./nbdkit --foreground \ --unix /tmp/nbd.sock \ --exportname '' \ --filter stats \ file file=/var/tmp/dst.img \ statsfile=/dev/stderr \ --run 'qemu-img

Previously we measured the total time and used it to calculate the rate of different operations. This is misleading and hiding the real throughput. Change single operation stats to use the time spent in this operation. To understand total system throughput, show also new "total" stats, summing ops, bytes of other operations, and using the total time. This works well for copying images

Change the first line to show total ops, time, total size, and rate. The totals are useful for understanding the total throughput of the application. --- filters/stats/nbdkit-stats-filter.pod | 8 ++++---- filters/stats/stats.c | 19 ++++++++++++++++++- 2 files changed, 22 insertions(+), 5 deletions(-) diff --git a/filters/stats/nbdkit-stats-filter.pod

On Mon, Jul 10, 2023 at 12:20:01PM +0300, Gavin Li wrote: > Currently, coalescing parameters are grouped for all transmit and receive > virtqueues. This patch series add support to set or get the parameters for > a specified virtqueue. > > When the traffic between virtqueues is unbalanced, for example, one virtqueue > is busy and another virtqueue is idle, then it will be very

On Thu, 13 Jul 2023 07:40:12 -0400, "Michael S. Tsirkin" <mst at redhat.com> wrote: > On Mon, Jul 10, 2023 at 12:20:01PM +0300, Gavin Li wrote: > > Currently, coalescing parameters are grouped for all transmit and receive > > virtqueues. This patch series add support to set or get the parameters for > > a specified virtqueue. > > > > When the

flush is usually called once but it can take significant time, so we need to include it. Here is an example run with this change: $ ./nbdkit --foreground \ --unix /tmp/nbd.sock \ --exportname '' \ --filter stats \ file file=/var/tmp/dst.img \ statsfile=/dev/stderr \ --run 'qemu-img convert -p -n -f raw -O raw -T none /var/tmp/fedora-30.img

I make some networking performance tests for Linux 3.10.13 #1 SMP is that normal Linux (pktget) ----> Linux (router) ----> Linux (Sink) eth0: 0.00 P/s 652024.78 P/s 652024.78 P/s eth1: 652175.58 P/s 2.98 P/s 652178.55 P/s with echo 0 >/proc/sys/net/bridge/bridge-nf-call-iptables CPU is about 2% with echo 1

This issue I am encountering seems to only happen on multi-socket machines. It also does not help that the only multi-socket box I have is an Romley-EP (so two socket SandyBridge CPUs). The other SandyBridge boxes I''ve (one socket) are not showing this. Granted they are also a different model (42). The problem is that when I run ''perf top'' within an SMP PVHVM guest,

This commit adds a formal specification of tvdiff_usec and a partial specification of subtract_timeval. These may be proved using Frama-C. The existing functions ignored overflow, but it is possible to call the functions with parameters that will cause overflow. So to create a formal specification I had to modify the functions to signal overflow. Luckily GCC and Clang have convenient

Make stats output easier to parse and more useful for humans. Here is an example output with this change: elapsed time: 1.33377 s read: 248 ops, 4.75 MiB, 3.56 MiB/s write: 78 ops, 32.64 MiB, 24.48 MiB/s trim: 33 ops, 1.00 GiB, 767.75 MiB/s flush: 9 ops, 0 bytes, 0 bytes/s --- filters/stats/nbdkit-stats-filter.pod | 11 ++++---- filters/stats/stats.c | 36

On Sat, Nov 30, 2019 at 02:17:05AM +0200, Nir Soffer wrote: > I find bytes and bits-per-second unhelpful and hard to parse. Using GiB > for sizes works for common disk images, and MiB/s works for common > storage throughput. > > Here is an example run with this change: > > $ ./nbdkit --foreground \ > --unix /tmp/nbd.sock \ > --exportname '' \ >

On Sun, Mar 5, 2023 at 11:49?PM Alvaro Karsz <alvaro.karsz at solid-run.com> wrote: > > Unify virtio_net_ctrl_coal_tx and virtio_net_ctrl_coal_rx structs into a > single struct, virtio_net_ctrl_coal, as they are identical. > > This patch follows the VirtIO spec patch: > https://lists.oasis-open.org/archives/virtio-comment/202302/msg00431.html > > Signed-off-by: Alvaro

Unify virtio_net_ctrl_coal_tx and virtio_net_ctrl_coal_rx structs into a single struct, virtio_net_ctrl_coal, as they are identical. This patch follows the VirtIO spec patch: https://lists.oasis-open.org/archives/virtio-comment/202302/msg00431.html Signed-off-by: Alvaro Karsz <alvaro.karsz at solid-run.com> --- drivers/net/virtio_net.c | 15 +++++++--------

--- tests/Makefile.am | 6 ++++ tests/test-info-conntime.sh | 65 +++++++++++++++++++++++++++++++++++ tests/test-info-time.sh | 68 +++++++++++++++++++++++++++++++++++++ tests/test-info-uptime.sh | 65 +++++++++++++++++++++++++++++++++++ 4 files changed, 204 insertions(+) diff --git a/tests/Makefile.am b/tests/Makefile.am index 5a65db7..abcce94 100644 --- a/tests/Makefile.am +++

Hi, questions will follow. --- diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c index 99f7bd4..13b3fb1 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bios.c +++ b/drivers/gpu/drm/nouveau/nouveau_bios.c @@ -40,8 +40,6 @@ #define BIOSLOG(sip, fmt, arg...) NV_DEBUG(sip->dev, fmt, ##arg) #define LOG_OLD_VALUE(x) //x -#define BIOS_USLEEP(n)

Either wallclock time, uptime or time since client connection can be reflected back to the client in a big endian binary structure. $ nbdkit reflection time --run 'nbdsh --connect $uri -c "sys.stdout.buffer.write(h.pread(12,0))" | hexdump -C' 00000000 00 00 00 00 5d 8f 24 c7 00 04 24 01 \ | / / $ date --date="@$(( 0x5d8f24c7 ))" Sat

Either wallclock time, uptime or time since client connection can be served back to the client in a big endian binary structure. $ nbdkit info time --run 'nbdsh --connect $uri -c "sys.stdout.buffer.write(h.pread(12,0))" | hexdump -C' 00000000 00 00 00 00 5d 8f 24 c7 00 04 24 01 └─────┬─────┘ ┌─┘ │ $ date