[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 Alan Modra changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED CC|amodra at gmail dot com| Resolution|--- |FIXED Assignee|unassigned at sourceware dot org |amodra at gmail dot com --- Comment #9 from Alan Modra --- Fixed. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 --- Comment #8 from cvs-commit at gcc dot gnu.org --- The master branch has been updated by Alan Modra : https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=f2032b672ac5f0c570f3a6d263ce6b764e7dca93 commit f2032b672ac5f0c570f3a6d263ce6b764e7dca93 Author: Alan Modra Date: Tue Nov 19 21:19:24 2019 +1030 PR24499, ignore --add-gnu-debuglink for archives objcopy --add-gnu-debuglink=foo.a.dbg foo.a just doesn't make any sense. Who puts executables in archives? PR 24499 * objcopy.c (copy_file): Ignore --add-gnu-debuglink for archives. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 --- Comment #7 from janisozaur --- I'm not saying it makes a lot of sense, it's simply allowed by objcopy and it's a case I run into every so often via makepkg: https://git.archlinux.org/pacman.git/tree/scripts/libmakepkg/tidy/strip.sh.in?id=0bc687e2e01f4e6e3f2a2f7ac10c7e1fdbaf5ba2#n69 (I haven't actually verified it is this particular call that triggers it, but there's only this one in whole repository, so chances are high). I will try working on verifying that's the root cause and possibly addressing it there (suggestions welcome), but as long as objcopy allows such call, I would like to address this here as well. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 --- Comment #6 from Alan Modra --- Archives are typically used as static libraries, that is, they contain a group of relocatable object files. I don't think that splitting off debug info to a separate archive makes sense for relocatable object files. You'd make the debug info unusable when linking the static library, and any gnu-debuglink notes will be dropped by the linker. So, what is in your foo.a that makes sense to objcopy --add-gnu-debuglink? -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 janisozaur changed: What|Removed |Added Attachment #11769|0 |1 is obsolete|| --- Comment #5 from janisozaur --- Created attachment 11783 --> https://sourceware.org/bugzilla/attachment.cgi?id=11783=edit objcopy: provide cache for gnu_debuglink's target CRC v3 Patch v3, now with passing test suite. Thanks for your comment, I will try checking why exactly objcopy gets called in such way. Regardless of that, I'd still like to get this patch reviewed and merged. While it might be odd, there is nothing preventing from such usage of objcopy and it gets a significant speedup. A similar patch has already landed in llvm-objcopy: https://reviews.llvm.org/D61343 -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 Alan Modra changed: What|Removed |Added CC||amodra at gmail dot com --- Comment #3 from Alan Modra --- I suggest running the binutils testsuite, which will show you that your patch is broken. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 --- Comment #2 from janisozaur --- It's been two weeks since I filed this ticket, the project is fairly active and I'm trying to fix a case that's of importance to me. I would appreciate _any_ kind of feedback on the patch I posted. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 janisozaur changed: What|Removed |Added Attachment #11754|0 |1 is obsolete|| --- Comment #1 from janisozaur --- Created attachment 11769 --> https://sourceware.org/bugzilla/attachment.cgi?id=11769=edit objcopy: provide cache for gnu_debuglink's target CRC v2 I've created a v2 of this patch, which I like better. Instead of relying on the variable caching value from possible previous iteration on an object, explicitly precompute the CRC and use it in all the cases. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24499] [PATCH] objcopy: provide cache for gnu_debuglink's target CRC
https://sourceware.org/bugzilla/show_bug.cgi?id=24499 janisozaur changed: What|Removed |Added CC||janisozaur+sourceware@gmail ||.com -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
