diff --git a/plugins/GSdx/GSDeviceOGL.cpp b/plugins/GSdx/GSDeviceOGL.cpp
index fd1c56f6f0..86902de7c3 100644
--- a/plugins/GSdx/GSDeviceOGL.cpp
+++ b/plugins/GSdx/GSDeviceOGL.cpp
@@ -1405,6 +1405,7 @@ void GSDeviceOGL::OMSetRenderTargets(GSTexture* rt, GSTexture* ds, const GSVecto
 	if(GLState::viewport != size)
 	{
 		GLState::viewport = size;
+		// FIXME ViewportIndexedf or ViewportIndexedfv (GL4.1)
 		glViewport(0, 0, size.x, size.y);
 	}
 
@@ -1413,6 +1414,7 @@ void GSDeviceOGL::OMSetRenderTargets(GSTexture* rt, GSTexture* ds, const GSVecto
 	if(!GLState::scissor.eq(r))
 	{
 		GLState::scissor = r;
+		// FIXME ScissorIndexedv (GL4.1)
 		glScissor( r.x, r.y, r.width(), r.height() );
 	}
 }
diff --git a/plugins/GSdx/GSTextureCache.cpp b/plugins/GSdx/GSTextureCache.cpp
index 67c139c0d4..4ee47a2a99 100644
--- a/plugins/GSdx/GSTextureCache.cpp
+++ b/plugins/GSdx/GSTextureCache.cpp
@@ -94,6 +94,14 @@ GSTextureCache::Source* GSTextureCache::LookupSource(const GIFRegTEX0& TEX0, con
 			continue;
 		}
 
+		// Special check for palette texture (psm.pal > 0)
+		//
+		// if m_paltex is enabled
+		// 1/ s->m_palette must always be defined
+		// 2/ Clut is useless (will be uploaded again at the end of the function)
+		//
+		// if m_paltex is disabled
+		// 1/ Clut must match if m_palette is NULL
 		if(s->m_palette == NULL && psm.pal > 0 && !GSVector4i::compare64(clut, s->m_clut, psm.pal * sizeof(clut[0])))
 		{
 			continue;
diff --git a/plugins/GSdx/GSTextureOGL.cpp b/plugins/GSdx/GSTextureOGL.cpp
index 26807cfcad..a1f4023d01 100644
--- a/plugins/GSdx/GSTextureOGL.cpp
+++ b/plugins/GSdx/GSTextureOGL.cpp
@@ -350,6 +350,7 @@ bool GSTextureOGL::Update(const GSVector4i& r, const void* data, int pitch)
 	g_real_texture_upload_byte += map_size;
 #endif
 
+	// PERF: slow path of the texture upload. Dunno if we could do better maybe check if TC can keep row_byte == pitch
 	// Note: row_byte != pitch
 	for (int h = 0; h < r.height(); h++) {
 		memcpy(map, src, row_byte);
diff --git a/plugins/GSdx/res/glsl/tfx_fs.glsl b/plugins/GSdx/res/glsl/tfx_fs.glsl
index fadb52640d..6fba02473e 100644
--- a/plugins/GSdx/res/glsl/tfx_fs.glsl
+++ b/plugins/GSdx/res/glsl/tfx_fs.glsl
@@ -262,8 +262,8 @@ vec4 sample_color(vec2 st, float q)
 	// PERF: see the impact of the exansion before/after the interpolation
 	for (int i = 0; i < 4; i++)
 	{
-        // PERF note: using dot produce reduces by 1 the number of instruction
-        // but I'm not it is equivalent neither faster.
+        // PERF note: using dot product reduces by 1 the number of instruction
+        // but I'm not sure it is equivalent neither faster.
         //float sum = dot(c[i].rgb, vec3(1.0f));
 #if ((PS_FMT & ~FMT_PAL) == FMT_24)
 		c[i].a = ( (PS_AEM == 0) || any(bvec3(c[i].rgb))  ) ? TA.x : 0.0f;
diff --git a/plugins/GSdx/res/glsl_source.h b/plugins/GSdx/res/glsl_source.h
index 77ab05a1df..f7a4d23153 100644
--- a/plugins/GSdx/res/glsl_source.h
+++ b/plugins/GSdx/res/glsl_source.h
@@ -1121,8 +1121,8 @@ static const char* tfx_fs_all_glsl =
 	"	// PERF: see the impact of the exansion before/after the interpolation\n"
 	"	for (int i = 0; i < 4; i++)\n"
 	"	{\n"
-	"        // PERF note: using dot produce reduces by 1 the number of instruction\n"
-	"        // but I'm not it is equivalent neither faster.\n"
+	"        // PERF note: using dot product reduces by 1 the number of instruction\n"
+	"        // but I'm not sure it is equivalent neither faster.\n"
 	"        //float sum = dot(c[i].rgb, vec3(1.0f));\n"
 	"#if ((PS_FMT & ~FMT_PAL) == FMT_24)\n"
 	"		c[i].a = ( (PS_AEM == 0) || any(bvec3(c[i].rgb))  ) ? TA.x : 0.0f;\n"