Re: [Mingw-w64-public] [PATCH] headers: Add ata.h

2021-10-08 Thread LIU Hao
在 2021-10-08 04:12, Mark Harmstone 写道: Signed-off-by: Mark Harmstone --- mingw-w64-headers/ddk/include/ddk/ata.h | 1732 +++ 1 file changed, 1732 insertions(+) create mode 100644 mingw-w64-headers/ddk/include/ddk/ata.h Thanks. Both patches look good to me. I applied

Re: [Mingw-w64-public] Patches to fix #515 and 916.

2021-10-08 Thread Martin Storsjö
On Fri, 8 Oct 2021, Martin Storsjö wrote: On Fri, 8 Oct 2021, David James wrote: Do you know when this will end up in a release of msys2? (And will/can I be notified somehow when it’s happened?) I think msys2 updates the mingw-w64-crt semi-sporadically, see e.g. this: https://github.com/msys

Re: [Mingw-w64-public] [PATCH] crt: Check pseudo relocations for overflows and error out clearly

2021-10-08 Thread Martin Storsjö
On Fri, 8 Oct 2021, LIU Hao wrote: 在 2021-10-08 17:36, Martin Storsjö 写道: Normally, when correctly configured, the pseudo relocations should be in fields that are large enough to hold the full target offset/address. But if the relocations nevertheless end up truncated, error out clearly instead

Re: [Mingw-w64-public] [PATCH 1/2] crt: Add APIs in computenetwork.def

2021-10-08 Thread LIU Hao
在 2021-10-08 02:09, Biswapriyo Nath 写道: From f381f33abc67c51626ff564155bfa7d9aabc8f7a Mon Sep 17 00:00:00 2001 From: Biswapriyo Nath Date: Thu, 7 Oct 2021 23:36:16 +0530 Subject: [PATCH 1/2] crt: Add APIs in computenetwork.def Signed-off-by: Biswapriyo Nath --- mingw-w64-crt/lib-common/comput

Re: [Mingw-w64-public] [PATCH] crt: Check pseudo relocations for overflows and error out clearly

2021-10-08 Thread LIU Hao
在 2021-10-08 17:36, Martin Storsjö 写道: Normally, when correctly configured, the pseudo relocations should be in fields that are large enough to hold the full target offset/address. But if the relocations nevertheless end up truncated, error out clearly instead of running into a hard to diagnose c

Re: [Mingw-w64-public] Patches to fix #515 and 916.

2021-10-08 Thread Martin Storsjö
On Fri, 8 Oct 2021, David James wrote: Do you know when this will end up in a release of msys2? (And will/can I be notified somehow when it’s happened?) I think msys2 updates the mingw-w64-crt semi-sporadically, see e.g. this: https://github.com/msys2/MINGW-packages/commits/master/mingw-w64-cr

Re: [Mingw-w64-public] Patches to fix #515 and 916.

2021-10-08 Thread David James
Great – thanks! (Sorry about the whitespace – not sure how that happened). Do you know when this will end up in a release of msys2? (And will/can I be notified somehow when it’s happened?) Thanks! David. From: Martin Storsjö Sent: 08 October 2021 10:42 To: mingw-w64-p

Re: [Mingw-w64-public] Patches to fix #515 and 916.

2021-10-08 Thread Martin Storsjö
On Thu, 7 Oct 2021, Martin Storsjö wrote: Hi, On Sun, 3 Oct 2021, David James wrote: Oops. Resending patches in .txt files. I’ve also reattached the test file, to which I added a small amount of additional testing. Thanks, David. Thanks for your patches - and it's great to have the corr

[Mingw-w64-public] [PATCH] crt: Check pseudo relocations for overflows and error out clearly

2021-10-08 Thread Martin Storsjö
Normally, when correctly configured, the pseudo relocations should be in fields that are large enough to hold the full target offset/address. But if the relocations nevertheless end up truncated, error out clearly instead of running into a hard to diagnose crash at runtime. The pseudo relocations

Re: [Mingw-w64-public] [PATCH] crt: Check pseudo relocations for overflows and error out clearly

2021-10-08 Thread Martin Storsjö
On Fri, 8 Oct 2021, LIU Hao wrote: 在 10/7/21 5:54 PM, Martin Storsjö 写道: + ptrdiff_t max_unsigned = (1LL << bits) - 1; + ptrdiff_t min_signed = (~(ptrdiff_t)0) << (bits - 1); The idea sounds correct. However it is undefined behavior to shift a negative number to the left,

Re: [Mingw-w64-public] [PATCH] crt: Check pseudo relocations for overflows and error out clearly

2021-10-08 Thread LIU Hao
在 10/7/21 5:54 PM, Martin Storsjö 写道: + ptrdiff_t max_unsigned = (1LL << bits) - 1; + ptrdiff_t min_signed = (~(ptrdiff_t)0) << (bits - 1); The idea sounds correct. However it is undefined behavior to shift a negative number to the left, so would you like to change the secon