MCAsmInfo: Allow targets to specify when the .section directive should be omitted

[• tstellarAMD]

Summary

The default behavior is to omit the .section directive for .text, .data,
and sometimes .bss, but some targets may want to omit this directive for
other sections too.

A future change to the AMDGPU backend will make use of this new feature
and have tests.

[• rafael]

I don't think this is correct. If a target is not using
COFF/MachO/ELF, it should never create a
MCSectionCOFF/MCSectionMacho/MCSectionELF.

IMHO what is needed is something like r245047 (with the logic for
picking the triple from -march fixed). With that you can them make
sure that you never create a section if your object format doesn't
have one.

[• tstellarAMD]

In D12423#235151, @rafael wrote:

I don't think this is correct. If a target is not using
COFF/MachO/ELF, it should never create a
MCSectionCOFF/MCSectionMacho/MCSectionELF.

This isn't the issue here. What this patch does is give targets the opportunity to control how section switches are emitted. Using AMDGPU as an example, if the section directive is not omitted, section switches to .hsatext are printed like this:

.section .hsatext,#alloc,#execinstr,#write

If the section directive is emitted switches are printed like this:

.hsatext

The main reason I don't want the .section directive for .hsatext sections is because it has several custom STT_* flags, and I didn't really see a good way to add support for printing these custom flags.

[• rafael]

Can you include that in the patch?

[• tstellarAMD]

Improved explanation of how shouldOmitSectionDirective() affects assembly output.

[• tstellarAMD]

Updated commit message to explain usage for AMDGPU backend. The new commit message looks like this:

The default behavior is to omit the .section directive for .text, .data,
and sometimes .bss, but some targets may want to omit this directive for
other sections too.

The AMDGPU backend will uses this to emit a simplified syntax for section
switches.  For example if the section directive is not omitted (current
behavior), section switches to .hsatext will be printed like this:

.section .hsatext,#alloc,#execinstr,#write

This is actually wrong, because .hsatext has some custom STT_* flags,
which MC doesn't know how to print or parse.

If the section directive is omitted (made possible by this commit),
section switches will be printed like this:

.hsatext

The motivation for this patch is to make it possible to emit sections
with custom STT_* flags without having to teach MC about all the target
specific STT_* flags.

[• rafael]

No, what I meant is to include a change to lib/Target/AMDGPU showing
this in use (and a test).

[• tstellarAMD]

This is the patch to make of use of this in the AMDPGU backend, and it also has tests: http://reviews.llvm.org/D12424

The AMDGPU patch is huge, but I can merge the two together if you want.

[• rafael]

Just noticed /D12424. Sorry for the confusion. LGTM for this one.

include/llvm/MC/MCAsmInfo.h

public:
   virtual const MCExpr *getExprForFDESymbol(const MCSymbol *Sym,
                                             unsigned Encoding,
                                             MCStreamer &Streamer) const;
 
   /// Return true if the identifier \p Name does not need quotes to be
   /// syntactically correct.
   virtual bool isValidUnquotedName(StringRef Name) const;
 
+  /// Return true if the .section directive should be omitted when
+  /// emitting \p SectionName.
+  virtual bool shouldOmitSectionDirective(StringRef SectionName) const;
+
   bool usesSunStyleELFSectionSwitchSyntax() const {
     return SunStyleELFSectionSwitchSyntax;
   }
 
   bool usesELFSectionDirectiveForBSS() const {
     return UsesELFSectionDirectiveForBSS;
   }
 

lib/MC/MCAsmInfo.cpp

bool MCAsmInfo::isValidUnquotedName(StringRef Name) const {
   // quotes.
   for (char C : Name) {
     if (!isAcceptableChar(C))
       return false;
   }
 
   return true;
 }
+
+bool MCAsmInfo::shouldOmitSectionDirective(StringRef SectionName) const {
+  // FIXME: Does .section .bss/.data/.text work everywhere??
+  return SectionName == ".text" || SectionName == ".data" ||
+        (SectionName == ".bss" && !usesELFSectionDirectiveForBSS());
+}

lib/MC/MCSectionELF.cpp

 // Decides whether a '.section' directive
 // should be printed before the section name.
 bool MCSectionELF::ShouldOmitSectionDirective(StringRef Name,
                                               const MCAsmInfo &MAI) const {
 
   if (isUnique())
     return false;
 
-  // FIXME: Does .section .bss/.data/.text work everywhere??
-  if (Name == ".text" || Name == ".data" ||
-      (Name == ".bss" && !MAI.usesELFSectionDirectiveForBSS()))
-    return true;
-
-  return false;
+  return MAI.shouldOmitSectionDirective(Name);
 }
 
 static void printName(raw_ostream &OS, StringRef Name) {
   if (Name.find_first_not_of("0123456789_."
                              "abcdefghijklmnopqrstuvwxyz"
                              "ABCDEFGHIJKLMNOPQRSTUVWXYZ") == Name.npos) {
     OS << Name;
     return;