WSL: ftruncate does not truncate mmap'd files #902

freeone3000 · 2016-08-15T20:24:33Z

Please use the following bug reporting template to help produce actionable and reproducible issues:

A brief description
ftruncate does not truncate mmap'd files while they are mmap'd.
Expected results
file gets truncated to desired size in bytes, even when mmap'd; SIGBUS occurs if memory outside new file bounds is accessed. (This is the behavior on Ubuntu.)
Actual results (with terminal output if applicable)
file does not get truncated
Your Windows build number
14393
Steps / commands required to reproduce the error

#include <sys/mman.h>
#include <fcntl.h>
#include <unistd.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <errno.h>
#define FAIL exit(errno);

int main() {
    /* prepare test data */
    int buf_size = 8102;
    char* buf = malloc(buf_size);
    if(NULL == buf) {
            FAIL
    }
    memset(buf, '?', buf_size);
    int fd = open("test_buffer", O_RDWR | O_CREAT);
    if(fd < 0) {
            FAIL
    }

    /* run test */
    write(fd, buf, buf_size);
    char* mm = mmap(NULL, buf_size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
    if(MAP_FAILED == mm) {
            FAIL
    }
    memset(mm, '!', buf_size);
    int tres = ftruncate(fd, 16);
    if(0 != tres) { /* okay, trying again later */
            write(STDOUT_FILENO, "NOFIRST\n", 9);
            errno = 0;
    }
    int mures = munmap(mm, buf_size);
    if(0 != mures) {
            FAIL
    }
    tres = ftruncate(fd, 16);
    if(0 != tres) {
            write(STDOUT_FILENO, "FAIL BOTH\n", 11);
            exit(1);
    }

    write(STDOUT_FILENO, "SUCCESS\n", 9);
    close(fd);
}

Strace of the failing command

  `execve("./min_test.bin", ["./min_test.bin"], [/* 23 vars */]) = 0
  brk(0)                                  = 0x12ab000
  access("/etc/ld.so.nohwcap", F_OK)      = -1 ENOENT (No such file or directory)
  mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f7c97540000
  access("/etc/ld.so.preload", R_OK)      = -1 ENOENT (No such file or directory)
  open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 3
  fstat(3, {st_mode=S_IFREG|0644, st_size=64852, ...}) = 0
  mmap(NULL, 64852, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7f7c97530000
  close(3)                                = 0
  access("/etc/ld.so.nohwcap", F_OK)      = -1 ENOENT (No such file or directory)
  open("/lib/x86_64-linux-gnu/libc.so.6", O_RDONLY|O_CLOEXEC) = 3
  read(3, "\177ELF\2\1\1\0\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0P \2\0\0\0\0\0"..., 832) = 832
  fstat(3, {st_mode=S_IFREG|0755, st_size=1840928, ...}) = 0
  mmap(NULL, 3949248, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f7c96e30000
  mprotect(0x7f7c96fea000, 2097152, PROT_NONE) = 0
  mmap(0x7f7c971ea000, 24576, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1ba000) = 0x7f7c971ea000
  mmap(0x7f7c971f0000, 17088, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7f7c971f0000
  close(3)                                = 0
  mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f7c97520000
  mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f7c97510000
  arch_prctl(ARCH_SET_FS, 0x7f7c97510740) = 0
  mprotect(0x7f7c971ea000, 16384, PROT_READ) = 0
  mprotect(0x600000, 4096, PROT_READ)     = 0
  mprotect(0x7f7c97422000, 4096, PROT_READ) = 0
  munmap(0x7f7c97530000, 64852)           = 0
  brk(0)                                  = 0x12ab000
  brk(0x12cd000)                          = 0x12cd000
  open("test_buffer", O_RDWR|O_CREAT, 0112547600) = 3
  write(3, "????????????????????????????????"..., 8102) = 8102
  mmap(NULL, 8102, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) = 0x7f7c9754b000
  ftruncate(3, 16)                        = -1 EINVAL (Invalid argument)
  write(1, "NOFIRST\n\0", 9)              = 9
  munmap(0x7f7c9754b000, 8102)            = 0
  ftruncate(3, 16)                        = 0
  write(1, "SUCCESS\n\0", 9)              = 9
  close(3)                                = 0
  exit_group(0)                           = ?
  +++ exited with 0 +++

Required packages and commands to install
apt-get -y install build-essential

See our contributing instructions for assistance.

The text was updated successfully, but these errors were encountered:

stehufntdev · 2016-08-16T19:17:17Z

Thanks for reporting the issue. This is discussed in #658 so closing it out as a duplicate. Please give us feedback on the user voice page so we can prioritize improving mmap support.

therealkenc · 2018-03-03T18:31:05Z

De-duping. The NOFIRST path still fires on 17110. Work around appears to be to munmap() the file like this.

gcc mmap-fail.c -o mmap-fail && rm -f test_buffer && strace -f -o mmap-fail.strace ./mmap-fail

WSL:

415   open("test_buffer", O_RDWR|O_CREAT, 0134077600) = 3
415   write(3, "????????????????????????????????"..., 8102) = 8102
415   mmap(NULL, 8102, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) = 0x7f48104f4000
415   ftruncate(3, 16)                  = -1 EINVAL (Invalid argument)
415   write(1, "NOFIRST\n\0", 9)        = 9
415   munmap(0x7f48104f4000, 8102)      = 0
415   ftruncate(3, 16)                  = 0
415   write(1, "SUCCESS\n\0", 9)        = 9

Real Linux:

25098 openat(AT_FDCWD, "test_buffer", O_RDWR|O_CREAT, 0171140) = 3
25098 write(3, "????????????????????????????????"..., 8102) = 8102
25098 mmap(NULL, 8102, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) = 0x7fe496235000
25098 ftruncate(3, 16)                  = 0
25098 munmap(0x7fe496235000, 8102)      = 0
25098 ftruncate(3, 16)                  = 0
25098 write(1, "SUCCESS\n\0", 9)        = 9

therealkenc · 2018-03-29T23:06:47Z

@stehufntdev @benhillis (Apologies for the ping) If this could be looked at on your end it will help put some indeterminate rpm / yum (and in general BerkeleyDB) issues to rest. We're going to get more users using non-Debian derived systems, and BerkeleyDB is used all over Creation. It is a clean repro up top on 17627.

ytrezq · 2018-03-30T01:06:55Z

I added a full repro. ʀᴘᴍ issues might not be linked to current case…

Uldiniad · 2018-05-03T00:37:07Z

any news on this issue?

drujd · 2018-05-26T10:38:35Z

I too would very much like to see this issue resolved. It blocks building Android/LineageOS on WSL:

[  2% 2392/93534] Generating TOC: /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc
FAILED: /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc
/bin/bash -c "(ASAN_OPTIONS=detect_leaks=0 prebuilts/build-tools/linux-x86/bin/ijar /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp ) && (if cmp -s /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc ; then rm /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp ; else mv /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc ; fi )"
ftruncate(fd_out, GetSize()): Invalid argument
/bin/bash: line 1: 26107 Aborted                 (core dumped) ( ASAN_OPTIONS=detect_leaks=0 prebuilts/build-tools/linux-x86/bin/ijar /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp )
[  2% 2397/93534] Generating TOC: /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc
FAILED: /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc
/bin/bash -c "(ASAN_OPTIONS=detect_leaks=0 prebuilts/build-tools/linux-x86/bin/ijar /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc.tmp ) && (if cmp -s /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc.tmp /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc ; then rm /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc.tmp ; else mv /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc.tmp /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc ; fi )"
ftruncate(fd_out, GetSize()): Invalid argument
/bin/bash: line 1: 26139 Aborted                 (core dumped) ( ASAN_OPTIONS=detect_leaks=0 prebuilts/build-tools/linux-x86/bin/ijar /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar /home/ddd/android/lineage/out/target/common/obj/JAVA_LIBRARIES/sdk_v17_intermediates/classes.jar.toc.tmp )
[  2% 2399/93534] Build hyb /home/ddd/android/lineage/out/target/product/gts210velte/obj/ETC/hyph-nb_intermediates/hyph-nb.hyb <- external/hyphenation-patterns/nb//hyph-nb.pat.txt
12585 unique nodes, 42861 total
[  2% 2400/93534] Build hyb /home/ddd/android/lineage/out/target/product/gts210velte/obj/ETC/hyph-nn_intermediates/hyph-nn.hyb <- external/hyphenation-patterns/nn//hyph-nn.pat.txt
12585 unique nodes, 42861 total
[  2% 2401/93534] Build hyb /home/ddd/android/lineage/out/target/product/gts210velte/obj/ETC/hyph-hu_intermediates/hyph-hu.hyb <- external/hyphenation-patterns/hu//hyph-hu.pat.txt
21515 unique nodes, 102669 total
ninja: build stopped: subcommand failed.
12:29:11 ninja failed with: exit status 1

It is also problematic with bup.

Uldiniad · 2018-05-26T12:54:38Z

@drujd check my xda thread to bypass the ijar issue: https://forum.xda-developers.com/android/software-hacking/guide-how-to-build-lineageos-15-1-t3750175

gao4263 · 2018-08-17T02:17:00Z

this issue on 16 Aug 2016 , why not fix ?

sirredbeard · 2019-02-25T23:43:35Z

Noting we have hit this issue in WLinux Enterprise on RHEL and Scientific Linux, specifically to some wonky rpm behavior here and there. See WhitewaterFoundry/Pengwin-Enterprise#20.

qclzdh · 2019-05-07T06:53:43Z

I encounter following error when build android with wsl on win10 18890, i also checked on google everywhere, on many page lead me to here, but i can't find any solution here, and also it seem this issue also solved in 18890, is there anyone can help me to solve this>
[ 0% 60/50136] Generating TOC: out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc FAILED: out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc /bin/bash -c "(ASAN_OPTIONS=detect_leaks=0 prebuilts/build-tools/linux-x86/bin/ijar out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp ) && (if cmp -s out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc ; then rm out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp ; else mv out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc ; fi )" ftruncate(fd_out, GetSize()): Invalid argument /bin/bash: line 1: 1072 Aborted (core dumped) ( ASAN_OPTIONS=detect_leaks=0 prebuilts/build-tools/linux-x86/bin/ijar out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp ) [ 0% 61/50136] Install: out/host/linux-x86/bin/aapt2

thorsteneb · 2019-05-07T11:26:10Z

If you haven’t seen it: https://www.howtogeek.com/413564/windows-10-is-getting-a-built-in-linux-kernel/

[deleted bits about WSL1 no longer being developed alongside wsl2 - I jumped the gun there]

I’m curious how the BerkeleyDB stuff is going to work since the Windows kernel doesn’t allow the kind of file system shenanigans that Linux does, and WSL files are available from both Windows and Linux. Maybe they’ll be kept in a lightweight VM now and just accessed from Windows.

My understanding is this kernel is the same or a very similar approach to what Azure does, so here’s to a fully compatible Linux experience on Windows without needing to spin up a full fat VM.

ytrezq · 2019-05-07T11:52:39Z

@yorickdowne : I still think this issue needs to be fixed. We’re not talking about a feature but the fact the implementation isn’t implementing something correctly.

There’s 3 points which are unclear with the new plan : will it remain possible to access Windows files like currently or cygwin ? Will it be required to manage a disk image it’s own file system with different sizes than the host drive (currently there’s no need for large images which mean free space of wsl or cygwin can be efficiently shared without having too much free space on image leaving no space on the host and vice versa). Will it be possible to still see each Linux process from Windows like with pico process currently ?

As currently phrased, this really sounds like leaving all those possibilities and switching to something with the same limitations as https://en.wikipedia.org/wiki/Cooperative_Linux (something far heavier than a Linux Docker). In that sense the plan to keep wsl1 really sounds to be a red flag for me indicating that huge benefits of the current implementation will be lost meaning Cygwin will remains superior in functionality over both implementations. Which in turns means this bug should be fixed (not being even able to run rpm as a single command but seriously).

drujd · 2019-05-07T17:02:22Z

@yorickdowne Nooooo! :(

So is it as I feared - WSL1 will stop being supported/developed?!

WSL2 is a glorified Linux VM. :( And, unless I'm wrong, it will require Hyper-V, therefore blocking (due to Microsoft's Hyper-V implementation) other virtualization providers, along with other problems stemming from running 'host' in a VM (albeit in a root partition).

drujd · 2019-05-07T17:04:54Z

@ytrezq Agreed, judging from the information available, WSL2 is a cop out compared to WSL1.

therealkenc · 2019-05-07T22:50:41Z

WSL1 will stop being supported/developed?!

The announcement goes out of its way to say: "Individual Linux distros can be run either as a WSL 1 distro, or as a WSL 2 distro, can be upgraded or downgraded at any time, and you can run WSL 1 and WSL 2 distros side by side."

Saying WSL1 will stop being supported, immediately anyway, is pretty overstated. For example:

I’m curious how the BerkeleyDB stuff is going to work

Actually it isn't clear whether BDB is still broken, since #3939 was squashed. Squashed just a little over a month ago. See 'cause the devs are still fixing issues in WSL1. Fixing issues at the same time the devs were frantically getting WSL2 in a state a screencap animated gif could be taken for Build 2019.

[BDB might still be broken in some subtle unreported way, mind. Dunno. Haven't looked. No well formed issue submission to look at, even for the curiosity value.]

Meanwhile this issue (#902) is still unresolved, of course. Says so at the top of the screen and everything.

There’s 3 points which are unclear with the new plan

There are lots of points about the new plan, both of the mostly clear and of the somewhat unclear variety. None of which have to do with this open issue specifically. The notable question we're faced with today is whether the WSL2 announcement implodes every favorite open issue into full chat mode for the next couple of months.

qclzdh · 2019-06-19T07:01:44Z

This issue was solved on wsl2, i can build android successfully now.

I encounter following error when build android with wsl on win10 18890, i also checked on google everywhere, on many page lead me to here, but i can't find any solution here, and also it seem this issue also solved in 18890, is there anyone can help me to solve this>
[ 0% 60/50136] Generating TOC: out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc FAILED: out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc /bin/bash -c "(ASAN_OPTIONS=detect_leaks=0 prebuilts/build-tools/linux-x86/bin/ijar out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp ) && (if cmp -s out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc ; then rm out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp ; else mv out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc ; fi )" ftruncate(fd_out, GetSize()): Invalid argument /bin/bash: line 1: 1072 Aborted (core dumped) ( ASAN_OPTIONS=detect_leaks=0 prebuilts/build-tools/linux-x86/bin/ijar out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar out/target/common/obj/JAVA_LIBRARIES/sdk_v8_intermediates/classes.jar.toc.tmp ) [ 0% 61/50136] Install: out/host/linux-x86/bin/aapt2

ytrezq · 2019-07-20T01:53:41Z

Agreed, judging from the information available, WSL2 is a cop out compared to WSL1.

@drujd Yikes ! It requires a file based ᴠʜᴅ Like any Windows Virtual Machine (the huge slow because of fileystem on hard drives because of the double layer of fragmentation and likely to make file system access slower than ᴡꜱʟ1). I would also add that I’ve more than 256Gb of code in my ᴡꜱʟ1 home directory simply barring access to ᴡꜱʟ2.
It’s definitely just a modern Microsoft Version of https://en.wikipedia.org/wiki/Cooperative_Linux

I also have many apps through otvdm which uses the Intel Haxm ᴀᴘɪ which is very unlikely to support Hyper-V.

Unless the ability to boot real Linux main rootfs from ᴡꜱʟ2 is given (boot from ʙᴛʀꜰꜱ or Linux ʟᴠᴍ or regular partition), this issue should definitely be fixed.

gordonliang · 2019-09-19T08:04:05Z

I got the same issue while compiling webrtc on WSL 1, I fix this issue with a workaround solution: munmap the mmap'd file before ftruncate and it's works fine for me.

microsoft/WSL#902

rubin55 · 2022-12-27T12:52:09Z

@therealkenc will this issue ever be fixed? I think a lot of users have specific + legitimate requirements to use WSL1 instead of WSL2; this issue has a lot of impact on various sofwares (re: the many tickets closed as duplicates to this one). Is this issue particularly hard to fix or won't MS give it priority? It makes WSL1 borderline unusable imho.. hoping hard for a fix.

Daniel-Khodabakhsh · 2023-03-15T00:50:18Z

@drujd check my xda thread to bypass the ijar issue: https://forum.xda-developers.com/android/software-hacking/guide-how-to-build-lineageos-15-1-t3750175

The link now shows a post that has been updated to only include WSL2 steps. But Waybackmachine is here to help: https://web.archive.org/web/20180617170133/https://forum.xda-developers.com/android/software-hacking/guide-how-to-build-lineageos-15-1-t3750175 however there's still some digging required.

Basically you need to apply this change to the synced ijar/zip.cc file: https://review.lineageos.org/c/LineageOS/android_build/+/208102. Make sure to make clean before re-attempting the brunch command.

microsoft-github-policy-service · 2024-03-14T12:58:30Z

This issue has been automatically closed since it has not had any activity for the past year. If you're still experiencing this issue please re-file this as a new issue or feature request.

Thank you!

microsoft-github-policy-service · 2024-03-14T12:58:30Z

This issue has been automatically closed since it has not had any activity for the past year. If you're still experiencing this issue please re-file this as a new issue or feature request.

Thank you!

drujd · 2024-03-14T13:02:11Z

So I guess that's it, the final confirmation (not that we needed one), WSL1 really is dead in the water, replaced by a glorified VM... :(

stehufntdev closed this as completed Aug 16, 2016

stehufntdev added the duplicate label Aug 16, 2016

therealkenc reopened this Mar 3, 2018

therealkenc removed the duplicate label Mar 3, 2018

therealkenc mentioned this issue Mar 3, 2018

mmap's problem #658

Closed

therealkenc mentioned this issue Mar 20, 2018

Segmentation fault when calling "rpm --initdb --dbpath" on Ubuntu for Windows #2871

Closed

therealkenc mentioned this issue Mar 29, 2018

rpmdb ‑‑initdb Database problem on Fedora because of a problem in libdb. Can’t install or remove any packages. #2852

Closed

therealkenc mentioned this issue May 5, 2018

Oracle Berkeley DB Files Issue #3149

Closed

This was referenced Aug 13, 2018

java.io.RandomAccessFile.setLength() failed when Kafka shutting down #2281

Closed

ijar's problem #3157

Closed

therealkenc mentioned this issue Aug 21, 2018

Fedora builds not working on WSL (dnf / yum failing) #3493

Closed

drujd mentioned this issue Aug 22, 2018

mmap(NULL, ...) on zero-length file fails with ENOEXEC #3451

Closed

fquinner mentioned this issue Aug 23, 2018

Unable to run Fedora 28,27 on Windows 10 v1803 DDoSolitary/LxRunOffline#46

Closed

therealkenc mentioned this issue Dec 23, 2018

BoltDB panics on cursor search since April update #3162

Closed

sirredbeard mentioned this issue Feb 25, 2019

rpm --rebuilddb causes all future rpm functions to segfault/yum to hang WhitewaterFoundry/Pengwin-Enterprise#20

Closed

sirredbeard mentioned this issue Mar 26, 2019

Unexpected behavior in mmap causing issues with rpm #3939

Closed

Dobatymo mentioned this issue May 17, 2019

on_disk_build bug spotify/annoy#353

Closed

PaulGrandperrin mentioned this issue Jun 28, 2019

Example Build Failed rust-fuzz/honggfuzz-rs#22

Closed

jnwatson mentioned this issue Oct 5, 2019

MDB_CORRUPTED error (Located page was wrong type) when trying to build database in WSL jnwatson/py-lmdb#215

Closed

IepIweidieng mentioned this issue Feb 11, 2020

Fix Pack 003: LUNATIC FORCE [Stage Final] ccns/dreambbs#61

Merged

minhbq mentioned this issue Mar 29, 2020

Error ftruncate (in unix.c[111]) when build and run libqb with WSL ClusterLabs/libqb#389

Closed

chrissie-c added a commit to chrissie-c/libqb that referenced this issue Oct 2, 2020

ipcs: ftruncate is not support on WIN32

3b8881b

microsoft/WSL#902

chrissie-c mentioned this issue Oct 2, 2020

ipcs: ftruncate is not support on WIN32 ClusterLabs/libqb#424

Merged

chrissie-c added a commit to ClusterLabs/libqb that referenced this issue Oct 5, 2020

ipcs: ftruncate is not support on WIN32 (#424)

51f64ed

microsoft/WSL#902

therealkenc added the wsl1 label May 5, 2021

qyb mentioned this issue Jul 15, 2022

run testrunner.pl under WSL1 cyrusimap/cyrus-imapd#4158

Open

microsoft-github-policy-service bot closed this as completed Mar 14, 2024

github-actions bot mentioned this issue Mar 14, 2024

WSL1: ftruncate does not truncate mmap'd files #11302

Open

2 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

WSL: ftruncate does not truncate mmap'd files #902

WSL: ftruncate does not truncate mmap'd files #902

freeone3000 commented Aug 15, 2016 •

edited by therealkenc

Loading

stehufntdev commented Aug 16, 2016

therealkenc commented Mar 3, 2018 •

edited

Loading

therealkenc commented Mar 29, 2018

ytrezq commented Mar 30, 2018

Uldiniad commented May 3, 2018

drujd commented May 26, 2018 •

edited

Loading

Uldiniad commented May 26, 2018

gao4263 commented Aug 17, 2018

sirredbeard commented Feb 25, 2019 •

edited

Loading

qclzdh commented May 7, 2019

thorsteneb commented May 7, 2019 •

edited

Loading

ytrezq commented May 7, 2019 •

edited

Loading

drujd commented May 7, 2019

drujd commented May 7, 2019

therealkenc commented May 7, 2019

qclzdh commented Jun 19, 2019 •

edited

Loading

ytrezq commented Jul 20, 2019 •

edited

Loading

gordonliang commented Sep 19, 2019

rubin55 commented Dec 27, 2022

Daniel-Khodabakhsh commented Mar 15, 2023

microsoft-github-policy-service bot commented Mar 14, 2024

microsoft-github-policy-service bot commented Mar 14, 2024

drujd commented Mar 14, 2024

WSL: ftruncate does not truncate mmap'd files #902

WSL: ftruncate does not truncate mmap'd files #902

Comments

freeone3000 commented Aug 15, 2016 • edited by therealkenc Loading

stehufntdev commented Aug 16, 2016

therealkenc commented Mar 3, 2018 • edited Loading

therealkenc commented Mar 29, 2018

ytrezq commented Mar 30, 2018

Uldiniad commented May 3, 2018

drujd commented May 26, 2018 • edited Loading

Uldiniad commented May 26, 2018

gao4263 commented Aug 17, 2018

sirredbeard commented Feb 25, 2019 • edited Loading

qclzdh commented May 7, 2019

thorsteneb commented May 7, 2019 • edited Loading

ytrezq commented May 7, 2019 • edited Loading

drujd commented May 7, 2019

drujd commented May 7, 2019

therealkenc commented May 7, 2019

qclzdh commented Jun 19, 2019 • edited Loading

ytrezq commented Jul 20, 2019 • edited Loading

gordonliang commented Sep 19, 2019

rubin55 commented Dec 27, 2022

Daniel-Khodabakhsh commented Mar 15, 2023

microsoft-github-policy-service bot commented Mar 14, 2024

microsoft-github-policy-service bot commented Mar 14, 2024

drujd commented Mar 14, 2024

freeone3000 commented Aug 15, 2016 •

edited by therealkenc

Loading

therealkenc commented Mar 3, 2018 •

edited

Loading

drujd commented May 26, 2018 •

edited

Loading

sirredbeard commented Feb 25, 2019 •

edited

Loading

thorsteneb commented May 7, 2019 •

edited

Loading

ytrezq commented May 7, 2019 •

edited

Loading

qclzdh commented Jun 19, 2019 •

edited

Loading

ytrezq commented Jul 20, 2019 •

edited

Loading