Metadata::created() not working on Linux with statx
#59743
Labels
C-enhancement
Category: An issue proposing an enhancement or a PR with one.
O-linux
Operating system: Linux
T-libs-api
Relevant to the library API team, which will review and decide on the PR/issue.
Comments
jonas-schievink
added
C-enhancement
Centril
added a commit
to Centril/rust
that referenced
this issue
Oct 9, 2019
Prefer statx on linux if available This PR make `metadata`-related functions try to invoke `statx` first on Linux if available, making `std::fs::Metadata::created` work on Linux with `statx` supported. It follows the discussion in rust-lang#61386 , and will fix rust-lang#59743 The implementation of this PR is simply converting `struct statx` into `struct stat64` with extra fields for `btime` if `statx` succeeds, since other fields are not currently used. --- I also did a separated benchmark for `fs::metadata`, `stat64`, `statx`, and `statx` with conversion to `stat64`. It shows that `statx` with conversion is even more faster than pure `statx`. I think it's due to `sizeof stat64 == 114` but `sizeof statx == 256`. Anyway, the bare implementation of `statx` with conversion is only about 0.2% slower than the original impl (`stat64`-family). With heap-allocation counted (~8.5% of total cost), the difference between `stat` and `statx` (with or without conversion) is just nothing. Therefore, I think it is not urgent to use bare `struct statx` as underlying representation now. There is no need to break `std::os::linux::fs::MetadataExt::as_raw_stat` (rust-lang#61386 (comment)) [Separated bare benchmarks](https://gist.github.com/oxalica/c4073ecb202c599fe41b7f15f86dc79c): ``` metadata_ok time: [529.41 ns 529.77 ns 530.19 ns] metadata_err time: [538.71 ns 539.39 ns 540.35 ns] stat64_ok time: [484.32 ns 484.53 ns 484.75 ns] stat64_err time: [481.77 ns 482.00 ns 482.24 ns] statx_ok time: [488.07 ns 488.35 ns 488.62 ns] statx_err time: [487.74 ns 488.00 ns 488.27 ns] statx_cvt_ok time: [485.05 ns 485.28 ns 485.53 ns] statx_cvt_err time: [485.23 ns 485.45 ns 485.67 ns] ``` r? @alexcrichton
Centril
added a commit
to Centril/rust
that referenced
this issue
Oct 15, 2019
Prefer statx on linux if available This PR make `metadata`-related functions try to invoke `statx` first on Linux if available, making `std::fs::Metadata::created` work on Linux with `statx` supported. It follows the discussion in rust-lang#61386 , and will fix rust-lang#59743 The implementation of this PR is simply converting `struct statx` into `struct stat64` with extra fields for `btime` if `statx` succeeds, since other fields are not currently used. --- I also did a separated benchmark for `fs::metadata`, `stat64`, `statx`, and `statx` with conversion to `stat64`. It shows that `statx` with conversion is even more faster than pure `statx`. I think it's due to `sizeof stat64 == 114` but `sizeof statx == 256`. Anyway, the bare implementation of `statx` with conversion is only about 0.2% slower than the original impl (`stat64`-family). With heap-allocation counted (~8.5% of total cost), the difference between `stat` and `statx` (with or without conversion) is just nothing. Therefore, I think it is not urgent to use bare `struct statx` as underlying representation now. There is no need to break `std::os::linux::fs::MetadataExt::as_raw_stat` (rust-lang#61386 (comment)) [Separated bare benchmarks](https://gist.github.com/oxalica/c4073ecb202c599fe41b7f15f86dc79c): ``` metadata_ok time: [529.41 ns 529.77 ns 530.19 ns] metadata_err time: [538.71 ns 539.39 ns 540.35 ns] stat64_ok time: [484.32 ns 484.53 ns 484.75 ns] stat64_err time: [481.77 ns 482.00 ns 482.24 ns] statx_ok time: [488.07 ns 488.35 ns 488.62 ns] statx_err time: [487.74 ns 488.00 ns 488.27 ns] statx_cvt_ok time: [485.05 ns 485.28 ns 485.53 ns] statx_cvt_err time: [485.23 ns 485.45 ns 485.67 ns] ``` r? @alexcrichton
Centril
added a commit
to Centril/rust
that referenced
this issue
Oct 16, 2019
Prefer statx on linux if available This PR make `metadata`-related functions try to invoke `statx` first on Linux if available, making `std::fs::Metadata::created` work on Linux with `statx` supported. It follows the discussion in rust-lang#61386 , and will fix rust-lang#59743 The implementation of this PR is simply converting `struct statx` into `struct stat64` with extra fields for `btime` if `statx` succeeds, since other fields are not currently used. --- I also did a separated benchmark for `fs::metadata`, `stat64`, `statx`, and `statx` with conversion to `stat64`. It shows that `statx` with conversion is even more faster than pure `statx`. I think it's due to `sizeof stat64 == 114` but `sizeof statx == 256`. Anyway, the bare implementation of `statx` with conversion is only about 0.2% slower than the original impl (`stat64`-family). With heap-allocation counted (~8.5% of total cost), the difference between `stat` and `statx` (with or without conversion) is just nothing. Therefore, I think it is not urgent to use bare `struct statx` as underlying representation now. There is no need to break `std::os::linux::fs::MetadataExt::as_raw_stat` (rust-lang#61386 (comment)) [Separated bare benchmarks](https://gist.github.com/oxalica/c4073ecb202c599fe41b7f15f86dc79c): ``` metadata_ok time: [529.41 ns 529.77 ns 530.19 ns] metadata_err time: [538.71 ns 539.39 ns 540.35 ns] stat64_ok time: [484.32 ns 484.53 ns 484.75 ns] stat64_err time: [481.77 ns 482.00 ns 482.24 ns] statx_ok time: [488.07 ns 488.35 ns 488.62 ns] statx_err time: [487.74 ns 488.00 ns 488.27 ns] statx_cvt_ok time: [485.05 ns 485.28 ns 485.53 ns] statx_cvt_err time: [485.23 ns 485.45 ns 485.67 ns] ``` r? @alexcrichton
Centril
added a commit
to Centril/rust
that referenced
this issue
Oct 16, 2019
Prefer statx on linux if available This PR make `metadata`-related functions try to invoke `statx` first on Linux if available, making `std::fs::Metadata::created` work on Linux with `statx` supported. It follows the discussion in rust-lang#61386 , and will fix rust-lang#59743 The implementation of this PR is simply converting `struct statx` into `struct stat64` with extra fields for `btime` if `statx` succeeds, since other fields are not currently used. --- I also did a separated benchmark for `fs::metadata`, `stat64`, `statx`, and `statx` with conversion to `stat64`. It shows that `statx` with conversion is even more faster than pure `statx`. I think it's due to `sizeof stat64 == 114` but `sizeof statx == 256`. Anyway, the bare implementation of `statx` with conversion is only about 0.2% slower than the original impl (`stat64`-family). With heap-allocation counted (~8.5% of total cost), the difference between `stat` and `statx` (with or without conversion) is just nothing. Therefore, I think it is not urgent to use bare `struct statx` as underlying representation now. There is no need to break `std::os::linux::fs::MetadataExt::as_raw_stat` (rust-lang#61386 (comment)) [Separated bare benchmarks](https://gist.github.com/oxalica/c4073ecb202c599fe41b7f15f86dc79c): ``` metadata_ok time: [529.41 ns 529.77 ns 530.19 ns] metadata_err time: [538.71 ns 539.39 ns 540.35 ns] stat64_ok time: [484.32 ns 484.53 ns 484.75 ns] stat64_err time: [481.77 ns 482.00 ns 482.24 ns] statx_ok time: [488.07 ns 488.35 ns 488.62 ns] statx_err time: [487.74 ns 488.00 ns 488.27 ns] statx_cvt_ok time: [485.05 ns 485.28 ns 485.53 ns] statx_cvt_err time: [485.23 ns 485.45 ns 485.67 ns] ``` r? @alexcrichton
Centril
added a commit
to Centril/rust
that referenced
this issue
Oct 17, 2019
Prefer statx on linux if available This PR make `metadata`-related functions try to invoke `statx` first on Linux if available, making `std::fs::Metadata::created` work on Linux with `statx` supported. It follows the discussion in rust-lang#61386 , and will fix rust-lang#59743 The implementation of this PR is simply converting `struct statx` into `struct stat64` with extra fields for `btime` if `statx` succeeds, since other fields are not currently used. --- I also did a separated benchmark for `fs::metadata`, `stat64`, `statx`, and `statx` with conversion to `stat64`. It shows that `statx` with conversion is even more faster than pure `statx`. I think it's due to `sizeof stat64 == 114` but `sizeof statx == 256`. Anyway, the bare implementation of `statx` with conversion is only about 0.2% slower than the original impl (`stat64`-family). With heap-allocation counted (~8.5% of total cost), the difference between `stat` and `statx` (with or without conversion) is just nothing. Therefore, I think it is not urgent to use bare `struct statx` as underlying representation now. There is no need to break `std::os::linux::fs::MetadataExt::as_raw_stat` (rust-lang#61386 (comment)) [Separated bare benchmarks](https://gist.github.com/oxalica/c4073ecb202c599fe41b7f15f86dc79c): ``` metadata_ok time: [529.41 ns 529.77 ns 530.19 ns] metadata_err time: [538.71 ns 539.39 ns 540.35 ns] stat64_ok time: [484.32 ns 484.53 ns 484.75 ns] stat64_err time: [481.77 ns 482.00 ns 482.24 ns] statx_ok time: [488.07 ns 488.35 ns 488.62 ns] statx_err time: [487.74 ns 488.00 ns 488.27 ns] statx_cvt_ok time: [485.05 ns 485.28 ns 485.53 ns] statx_cvt_err time: [485.23 ns 485.45 ns 485.67 ns] ``` r? @alexcrichton
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
First of all, it now returns an Err with "not available on this platform currently".
But I found that there is a system call
statxavailable in Linux kernel 4.11, which can get the birth time.As the doc, the function uses
staton Unix platforms. But thestatresult just never containsbirthtime.Shouldn't it try
statxwhen available?The text was updated successfully, but these errors were encountered: