int main()
{
for (int i = 0; i < (1 << 16); i++) {
void *p = mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_32BIT, -1, 0);
printf("%p\n", p);
if (p == (void *)-1) {
printf("%d\n", i);
return 0;
}
}
}
```
Test:
```nohighlight
$ uname -srvmo
Linux 6.3.1-arch1-1 #1 SMP PREEMPT_DYNAMIC Mon, 01 May 2023 17:42:39 +0000 x86_64 GNU/Linux
$ gcc -o mmap-test mmap-test.c
$ ./mmap-test
0x40898000
...
0x417e2000
0xffffffffffffffff
23
```
---
### Author:  [@fommil](https://discourse.haskell.org/u/fommil)
#### Post date: [May 11, 2023, 8:22am UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/8 "2023-05-11T08:22:31Z")
Unfortunately, upgrading to ghc 9.4 isn’t an option for many, including those of us who need to check that library code is compiling on older versions of ghc.
I’m experiencing this with both the current and LTS versions of the linux kernel on ArchLinux (6.3.1-arch2-1 and 6.1.27-1). Do we know if the upstream kernel is going to be fixing this promptly or should we be looking into compiling our own kernels and running them (or downgrading)? 6.2.9 works for me.
---
### Author:  [@fommil](https://discourse.haskell.org/u/fommil)
#### Post date: [May 11, 2023, 8:25am UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/9 "2023-05-11T08:25:39Z")
using
`export GHCRTS='-xp'`
gives me a
```haskell
R_X86_64_32S relocation out of range: (noname) = 7fdf70318050
```
on ghc-8.10.7 compiled apps, vs
` mmap 4964352 bytes at (nil): Cannot allocate memory` when I don’t use any flags.
---
### Author:  [@janus](https://discourse.haskell.org/u/janus)
#### Post date: [May 12, 2023, 9:13pm UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/10 "2023-05-12T21:13:43Z")
> [@fommil](#):
>
> Do we know if the upstream kernel is going to be fixing this promptly
Liam Howlett just mailed in 35 patches that seem related: ['[PATCH v3 00/35] Maple tree mas\_{next,prev}\_range() and cleanup' - MARC](https://marc.info/?l=linux-mm&m=168391557824417&w=2)
---
### Author:  [@janus](https://discourse.haskell.org/u/janus)
#### Post date: [May 14, 2023, 3:42pm UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/11 "2023-05-14T15:42:37Z")
> [@fommil](#):
>
> I’m experiencing this with […] 6.1.27-1
[Robert Hensing reports](https://lore.kernel.org/lkml/190c5b1b-df67-eb54-1559-706ed4e5719a@hercules-ci.com/) (possibly using bisect?) that an issue was reintroduced into the 6.1 series in commit [58c5d0d](https://github.com/torvalds/linux/commit/58c5d0d6d522112577c7eeb71d382ea642ed7be4), titled `mm/mmap: regression fix for unmapped_area{_topdown}`. Googling this commit, I found it in the [6.1.26 changelog](https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.26) and the [6.3 changelog](https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.3).
So it’s likely that 6.1.25 is ok.
> [@gilmi](#):
>
> I think it was even fixed in 6.3.0-rc7 but for some reason not in 6.3.1.
The title of commit 58c5d0d was [not listed in the rc7 changelog](https://lwn.net/Articles/929226/) (compare [6.3 announcement](https://lwn.net/Articles/929943/), where it is present). So this is consistent with your observation.
---
### Author:  [@juhp](https://discourse.haskell.org/u/juhp)
#### Post date: [May 18, 2023, 6:10am UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/12 "2023-05-18T06:10:27Z")
Unfortunately the ghci 9.2 crash is happening sometimes in current Fedora releases now (~10% of the time in kernel-6.2.15-300; 6.2.9-300 is okay) and more often with kernel-6.3.3-200 and 6.4.0 rc’s (rawhide).
So I am starting to wonder if this behavior will really change (back) in the Linux kernel or not?
(Though I am hoping to ship ghc-9.4 in Fedora 39…)
The crash happens for me only sometimes and before ghci even shows the initial prompt
(the core files are about 4MB apparently).
Also worth mentioning Sylvain’s [draft backport MR](https://gitlab.haskell.org/ghc/ghc/-/merge_requests/10453) too (not a small patch though by any definition). If you are using Fedora 38 you can update to this [temporary scratch build](https://koji.fedoraproject.org/koji/taskinfo?taskID=101285362) which includes the patches, eg using `koji-tool install 101285362`.
---
### Author:  [@fommil](https://discourse.haskell.org/u/fommil)
#### Post date: [May 18, 2023, 9:54am UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/13 "2023-05-18T09:54:29Z")
I don’t know if all those patches are included, but I can confirm that 6.3.2 does NOT fix the problem. I’m holding onto 6.2.9 for now.
---
### Author:  [@juhp](https://discourse.haskell.org/u/juhp)
#### Post date: [May 18, 2023, 3:49pm UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/14 "2023-05-18T15:49:53Z")
Is there a minimal reproducer in Haskell?
---
### Author:  [@tcard](https://discourse.haskell.org/u/tcard)
#### Post date: [May 18, 2023, 11:50pm UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/15 "2023-05-18T23:50:14Z")
I am working on a [Reflex FRP](https://reflex-frp.org/) project, where upgrading to GHC 9.4 is not an option. Using `export GHCRTS='-xp'` causes [Obelisk](https://github.com/obsidiansystems/obelisk#readme) to fail, reporting that it does not support that RTS option, but adding `+RTS -xp -RTS` to [`baseGhciOptions`](https://github.com/obsidiansystems/obelisk/blob/6125e83cb7eee8b5fbd5a4694db85fee9f286c6f/lib/command/src/Obelisk/Command/Run.hs#L549-L555) as follows makes the `ob run` and `ob repl` commands work for me. Thank you @fintara for the hint about using that RTS option!
```diff
diff --git a/lib/command/src/Obelisk/Command/Run.hs b/lib/command/src/Obelisk/Command/Run.hs
index 0e2b6b3d..6fce8551 100644
--- a/lib/command/src/Obelisk/Command/Run.hs
+++ b/lib/command/src/Obelisk/Command/Run.hs
@@ -552,6 +552,7 @@ baseGhciOptions =
, "-no-user-package-db"
, "-hide-all-packages"
, "-package-env", "-"
+ , "+RTS", "-xp", "-RTS"
]
-- | Run ghci repl
```
For Nix builds that have caused problems, I am simply doing those in a VM running an older kernel.
```nohighlight
$ uname -a
Linux ghc810 5.15.110 #1-NixOS SMP Sun Apr 30 23:23:24 UTC 2023 x86_64 GNU/Linux
```
---
### Author:  [@wiz](https://discourse.haskell.org/u/wiz)
#### Post date: [May 19, 2023, 12:49pm UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/16 "2023-05-19T12:49:59Z")
Just checked 6.3.3, no good.
---
### Author:  [@fommil](https://discourse.haskell.org/u/fommil)
#### Post date: [May 25, 2023, 4:15pm UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/17 "2023-05-25T16:15:18Z")
this is now fixed (again) in linux kernel `6.3.4`
---
### Author:  [@juhp](https://discourse.haskell.org/u/juhp)
#### Post date: [May 26, 2023, 6:42am UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/18 "2023-05-26T06:42:30Z")
Thanks @fommil
The maple\_tree fix also appears to be in Linux kernel [6.4-rc3](https://lore.kernel.org/lkml/CAHk-=wgL4Sa64qUZkpAs06t9UKoNyAYqBpzmukefJz8P8sa2=Q@mail.gmail.com/), so that is also good news.
I tested it too successfully already in Fedora Rawhide.
Also I have pushed Fedora updates for the ghc and ghc9.2 packages with the backport into testing.
---
### Author:  [@mpickering](https://discourse.haskell.org/u/mpickering)
#### Post date: [May 26, 2023, 3:03pm UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/19 "2023-05-26T15:03:28Z")
We have released [ghc-9.2.8](http://discourse.haskell.org/t/ghc-9-2-8-is-now-available/6328) with a fix to this issue.
---
### Author:  [@markus](https://discourse.haskell.org/u/markus)
#### Post date: [July 6, 2023, 9:40am UTC](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259/21 "2023-07-06T09:40:10Z")
For anyone still stuck on a ghc prior to 9.2.8, adding the following to your cabal component works:
```haskell
ghc-options:
+RTS -xp -RTS
```
You could do `cabal repl --ghc-options='+RTS -xp -RTS' some-target` but that will also rebuild dependencies.
[Next page](https://discourse.haskell.org/t/facing-mmap-4096-bytes-at-nil-cannot-allocate-memory-youre-not-alone/6259.md?page=2)