From 6458a8ac2baaf15308b1ba5d3d77c9208fe09886 Mon Sep 17 00:00:00 2001
From: Eric Long <i@hack3r.moe>
Date: Tue, 31 Oct 2023 18:23:01 +0800
Subject: [PATCH] lmdb: Platform-specific default map size

The current code works fine on x86 platforms since x86-64 uses at least
48-bit of virtual address space. On other 64-bit platforms like aarch64
or riscv64, the minimum allowed virtual address is 39-bit [1] [2].
Current 2**40 allocation will fail:

```
zict/tests/test_lmdb.py:53:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

self = <zict.lmdb.LMDB object at 0x40043bfc10>
directory = '/tmp/test_lmdb-sfepfnw8'

    def __init__(self, directory: str):
        import lmdb

        # map_size is the maximum database size but shouldn't fill up the
        # virtual address space
        map_size = 1 << 40 if sys.maxsize >= 2**32 else 1 << 28
        # writemap requires sparse file support otherwise the whole
        # `map_size` may be reserved up front on disk
        writemap = sys.platform.startswith("linux")
>       self.db = lmdb.open(
            directory,
            subdir=True,
            map_size=map_size,
            sync=False,
            writemap=writemap,
        )
E       lmdb.Error: /tmp/test_lmdb-sfepfnw8: Operation not supported

zict/lmdb.py:43: Error
```

Switching to 2**37 on aarch64 and riscv64 should fix the issue.

[1]: https://www.kernel.org/doc/html/v5.8/arm64/memory.html
[2]: https://www.kernel.org/doc/html/latest/riscv/vm-layout.html
---
 zict/lmdb.py | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/zict/lmdb.py b/zict/lmdb.py
index 71b1005..8ff4dc7 100644
--- a/zict/lmdb.py
+++ b/zict/lmdb.py
@@ -2,6 +2,7 @@
 
 import pathlib
 import sys
+import platform
 from collections.abc import ItemsView, Iterable, Iterator, ValuesView
 
 from zict.common import ZictBase
@@ -25,7 +26,8 @@ class LMDB(ZictBase[str, bytes]):
     directory: str
     map_size: int
         On Linux and MacOS, maximum size of the database file on disk.
-        Defaults to 1 TiB on 64 bit systems and 1 GiB on 32 bit ones.
+        Defaults to 128 GiB on aarch64 and riscv64, 1 TiB on other 64 bit systems like
+        x86-64 and 1 GiB on 32 bit ones.
 
         On Windows, preallocated total size of the database file on disk. Defaults to
         10 MiB to encourage explicitly setting it.
@@ -48,10 +50,17 @@ def __init__(self, directory: str | pathlib.Path, map_size: int | None = None):
 
         super().__init__()
         if map_size is None:
-            if sys.platform != "win32":
-                map_size = min(2**40, sys.maxsize // 4)
-            else:
+            machine = platform.machine()
+            if sys.platform == "win32":
                 map_size = 10 * 2**20
+            elif machine in ["x86_64", "x64"]:
+                map_size = 2**40
+            elif machine in ["i386", "i686", "x86"]:
+                map_size = 2**30
+            elif machine.startswith("aarch64") or machine.startswith("armv8") or machine.startswith("riscv64"):
+                map_size = 2**37
+            else:
+                map_size = min(2**40, sys.maxsize // 4)
 
         self.db = lmdb.open(
             str(directory),